Classification of mesoscopic

tribological properties under dry sliding

friction for microforming operation
Tetsuhide Shimizu & Ming Yang &
Ken-Ichi Manabe
Visit to download the full and correct content document:
https://ebookmass.com/product/classification-of-mesoscopic-tribological-properties-un
der-dry-sliding-friction-for-microforming-operation-tetsuhide-shimizu-ming-yang-ken-ic
hi-manabe/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Elsevier Weekblad - Week 26 - 2022 Gebruiker

https://ebookmass.com/product/elsevier-weekblad-
week-26-2022-gebruiker/

Jock Seeks Geek: The Holidates Series Book #26 Jill

Brashear

https://ebookmass.com/product/jock-seeks-geek-the-holidates-
series-book-26-jill-brashear/

The New York Review of Books – N. 09, May 26 2022

Various Authors

https://ebookmass.com/product/the-new-york-review-of-
books-n-09-may-26-2022-various-authors/

Fatigue and Tribological Properties of Plastics and

Elastomers, Third Edition Mckeen

https://ebookmass.com/product/fatigue-and-tribological-
properties-of-plastics-and-elastomers-third-edition-mckeen/
Calculate with Confidence, 8e (Oct 26,
2021)_(0323696953)_(Elsevier) 8th Edition Morris Rn
Bsn Ma Lnc

https://ebookmass.com/product/calculate-with-
confidence-8e-oct-26-2021_0323696953_elsevier-8th-edition-morris-
rn-bsn-ma-lnc/

Optimal Operation of Integrated Multi-Energy Systems

Under Uncertainty 1st Edition Wu

https://ebookmass.com/product/optimal-operation-of-integrated-
multi-energy-systems-under-uncertainty-1st-edition-wu/

An experimental investigation of the effects of the

compliant joint method on feedback compensation of pre-
sliding/pre-rolling friction Xin Dong & Chinedum E.
Okwudire
https://ebookmass.com/product/an-experimental-investigation-of-
the-effects-of-the-compliant-joint-method-on-feedback-
compensation-of-pre-sliding-pre-rolling-friction-xin-dong-
chinedum-e-okwudire/

Standard Penetration Tests and Soil Liquefaction

Potential Evaluation Fumio Tatsuoka & Toshio Iwasaki &
Ken-Ichi Tokida & Susumu Yasuda & Makoto Hirose &
Tsuneo Imai & Masashi Kon-No
https://ebookmass.com/product/standard-penetration-tests-and-
soil-liquefaction-potential-evaluation-fumio-tatsuoka-toshio-
iwasaki-ken-ichi-tokida-susumu-yasuda-makoto-hirose-tsuneo-imai-
masashi-kon-no/

Dry Syngas Purification Processes for Coal Gasification

Systems Makoto Kobayashi

https://ebookmass.com/product/dry-syngas-purification-processes-
for-coal-gasification-systems-makoto-kobayashi/
 Wear 330-331 (2015) 49–58

Contents lists available at ScienceDirect

Wear
journal homepage: www.elsevier.com/locate/wear

Classification of mesoscopic tribological properties under dry sliding

friction for microforming operation
Tetsuhide Shimizu a,n, Ming Yang a, Ken-ichi Manabe b
a
Division of Human Mechatronics Systems, Graduate School of System Design, Tokyo Metropolitan University, Japan 6-6, Asahigaoka, Hino-shi,
191-0065 Tokyo, Japan
b
Department of Mechanical Engineering, Graduate School of Science and Engineering, Tokyo Metropolitan University, Japan 1-1 Minamiohsawa,
Hachioji-shi, 192-0397 Tokyo, Japan

art ic l e i nf o a b s t r a c t

Article history: To generally describe the size effect in microforming operation under dry friction, which strongly
Received 6 September 2014 depends on the contact state and geometrical/mechanical properties of the surface, the present study
Received in revised form proposes a classification approach based on the theoretical scale-dependent model in microtribology.
15 January 2015
The proposed approach was applied to the practical problem of a scaled foil strip drawing test involving
Accepted 24 January 2015
contact between a pair of compression tools made of WC–Co hard alloy and a phosphor bronze foil with
a thickness of 20, 50, or 100 μm. From the results, it was clarified that the decreasing tendency of the
Keywords: total friction coefficient in the scaled foil strip drawing test was mainly due to the size effect of three-
Microforming body deformation (plowing), which was promoted by the generation of wear particles. The applicability
Dry sliding friction
of the proposed method and the positioning in the transitional scale range of mesoscopic tribology in
Size effect
microforming operation were quantitatively demonstrated.
Strip drawing test
& 2015 Elsevier B.V. All rights reserved.

1. Introduction
With the trend towards higher integrated multifunctionality and
the miniaturization of devices, further improvement of the forming
accuracy in microforming, which is known as a miniaturized
process based on conventional metal stamping technology, is
strongly required by the industry [1]. In particular, since the surface
area to volume ratio greatly increases with the miniaturization of
process dimensions, the interfacial behavior between forming dies
and work materials has an increasingly strong effect on formability
and its accuracy. However, owing to the so-called size effect, the
process design in microforming faces various problems in the
further development of tool design and process control [2].
Over the last decade, basic studies on the size effect of tribology in
microforming have been performed worldwide. The general under-
standing regarding the size effect of friction is the low effect of a
lubricant in the micro-scale forming process [3,4]. Although this low
effect of a lubricant is well described by the lubricant pocket model
proposed by Engel [5], the scale dependence of microforming opera-
tion under dry friction has not been well discussed. Basic investiga-
tions of the size effect in the case of dry friction have only been carried
out for bulk metal forming [6–9]. However, the conclusions regarding
the scale dependence in the case of the dry friction are still too
n
Corresponding author. Tel.: þ 81 42 585 8650; fax: þ 81 42 585 5119.
E-mail address: simizu-tetuhide@tmu.ac.jp (T. Shimizu).
ambiguous to determine the size effect; some authors found that there
was no size effect in scaled upsetting [6] and scaled forward extrusion
[7], while others stated that the friction increased in scaled ring
compression tests [8] or decreased in scaled compression tests [9]
with decreasing process dimensions. This is clearly due to the specific
contact states in each experiment, such as the contact pressure or the
surface properties of the work materials and forming die.
Since the predominant factors in the case of the dry friction are
the adhesion and deformation resulting from the direct interaction
of surface asperity peaks in contact with each other, the mechan-
ical property and geometry of the surface asperities directly affect
the friction and wear behavior under dry friction [10]. Additionally,
as a result of the deformation of asperities, wear debris of various
sizes is generated, and the trapped particles at the interface induce
the three-body plowing behavior, which significantly increases the
friction resistance and wear rate [11]. Therefore, the existence of
the size effect under dry friction strongly depends on the contact
state and surface properties. Thus, to generally describe the size
effect under the dry friction by a coherent analytical method,
classification of the scale dependence based on the contact state,
and the geometrical and mechanical properties of the surface is
required. In particular, since microforming operation deals with
the tribological behavior at the mesoscopic scale, which can be
positioned as the transitional area from micro- to macro- tribol-
ogy, identification of whether the tribology is scale dependent or
independent is of great significance. However, only a few dry
friction models have been proposed for microforming operation

http://dx.doi.org/10.1016/j.wear.2015.01.050
0043-1648/& 2015 Elsevier B.V. All rights reserved.
50 T. Shimizu et al. / Wear 330-331 (2015) 49–58

Nomenclature Ra Arithmetic average surface roughness

Rn Composite roughness of contact pair
List of symbols
rcorner Corner radius of the compression tool in strip
drawing test
Aa, Ad, Ap Real area of contact contributed to each component of rk Autocorrelation function
friction coefficient s Spacing between individual slip steps
Arp, Arp0 Real area of contact (Plastic contact) and its macro t, t0 Sheet thickness and its initial value
scale value Wad Surface energy of adhesion between contact pair
Are, Are0 Real area of contact (Elastic contact) and its macro W Applied load
scale value xscan Scanned position in surface profile measurement
a Contact radius
b Burgers vector
Greek letters
D Thickness of the finite area of dislocation density at
interface
Dp Index of penetration depth
α Probability of leaving of wear particle from the contact
surface
Dslide Sliding distance in scaled strip drawing test
d Size of the wear particles
βn, βn0 Correlation length of surface profile and its macro
scale value
d Average size of trapped wear particle
E, E1, E2 Young's modulus, and that for strip foils and
β Curvature radius of surface asperities
compression tools
η Density of the number of wear particles
En Effective elastic modulus of contact surface
θ Average surface roughness angle of harder material of
contact pair
FN Normal compression force in strip drawing test
FT Tensile force in strip drawing test
λ Scaling factor
Fμ Friction force in strip drawing test
μ, μtotal Friction coefficient, total coefficient of friction
Fa, Fd, Fp Friction force contributed to each component of fric-
μa Adhesion component of friction coefficient
tion coefficient
μd Deformation component of friction coefficient
H Hardness of softer material of contact pair
μp Plowing component of friction coefficient
hr Roughness height
μr Ratchet component of friction coefficient
k Autocorrelation interval
μs Static friction coefficient
L Characteristic length of contact problem
μk Kinetic friction coefficient
Llc Long wavelength limit of contact surface geometry
ν1, 2 Poisson's ratio of contact pair (Strip foils and
compression tools)
Llwl Long wavelength limit of surface geometry
Ld Characteristic length parameter related to ld
ρ Mass density
ld Plastic deformation length
ρG Density of GND (Geometrically necessary dislocation)
ls Average distance dislocation length
ρS Density of SSD (Statistically dtored dislocations)
m, n Multiplier index of the empirical rule of scale depen-
σY Yield stress
dence of surface geometry
σ, σ0 Standard deviation of height of surface profile and its
macro scale value
ntr Ratio of the number for trapped wear particles to the
total number of particles
τY Yield shear strength
pa Apparent surface pressure
τa, τd, τp, Shear strength contributed to each component of
friction coefficient
p (d) Probability density function of wear particles
ptr(d) Probability density function of trapped wear particles
Ψ Plasticity index
pac Critical apparent surface pressure which generates the
scale dependency

[12], and there have been no studies to generally classify the scale
dependence for dry sliding friction by a theoretical approach. To
construct guidelines for design based on lubricant-free micro-
forming operation, which is strongly required by industry [13], the
relevant parameters and scale effects under the contact state in
microforming must be specified.
Within the above background, the aim of the present study is to
characterize the mesoscopic tribological properties in the case of
dry sliding friction by classifying the size effect based on a
theoretical approach that considers surface asperities and three-
body deformation. A classification method that is based on a scale-
dependent model of dry friction in microtribology for MEMS
(micro electro mechanical systems) application is proposed in
Section 3. In Section 4, this proposed analytical method of
classification is applied to practical experimental data obtained
in the scaled strip drawing tests reported in [14,15], which are
briefly summarized in Section 2.
2. Experimental
One of the simplest but most practical testing methods for the
identification of tribological behavior in a conventional sheet
metal forming process is the strip drawing test. In the strip
drawing process, a compressed strip clamped between two
compression tools slides between the tools, generating frictional
forces (Fμ1, Fμ2) that oppose the movement of the strip, as shown
schematically in Fig. 1. Since the contributions of the friction forces
to the drawing force (FT) are nearly equal, this testing method
enables the semi-direct measurement of the friction coefficient.
Another advantage of this method is the ease of the parameter
variation, which exist the hundreds of parameters for the tribolo-
gical behavior, such as the scale dimensions and surface properties
of the strip material and compression tools, the normal force, and
the sliding speed etc. A number of studies on the strip drawing test
have provided significant results in the field of process tribology
 T. Shimizu et al. / Wear 330-331 (2015) 49–58
for conventional sheet metal forming, e.g. tribological character-
ization against the several coatings during deep drawing [16], pit
evolution of sheet metal during rolling [17], effect of transfer layer
[18], die surface patterning [19], and hard coatings (W–Ti–N) films
[20] during sheet metal forming.
To investigate the scale dependence of the friction behavior
under dry friction, scaled strip drawing tests were carried out while
maintaining geometrically similar conditions, by using the devel-
oped foil strip drawing tester [14]. The initial thicknesses of the foil
specimens were 20, 50, and 100 μm with a fixed ratio of the foil
thickness to the corner radius of the compression tools of t0/r0. As
an index for downscaling, a scaling factor λ was defined. In the
present experiment, λ was set to 1 for the 100 μm thick foil. The
strip samples were made of phosphor bronze (JIS: C5191-H) foils.
Figs. 2 and 3 show surface images of each foil and those at the
drawing tool corner radius for each scale dimension. These differ-
ences in surface roughness for different scale factors are also
considered in the theoretical analysis in Section 4. All the compres-
sion tools were made of sintered WC–Co hard alloy (JIS: V20
tungsten–carbide–cobalt alloy). All the parameters input to describe
Tensile force
FT 2Fμ
2μ·FN Foil strip specimen
Compression Compression
force FN force FN
Compression tools
Friction force Fμ
Fig. 1. Schematic illustration of foil strip drawing test showing the relations
between the applied forces.
TD
RD 20μm
Fig. 2. Surface images of strip specimens with thicknesses of (a) 20 μm, (b) 50 μm, and (c) 100 μm obtained by laser scanning microscopy.
Drawing direction
20μm
Fig. 3. Surface images at corner of drawing tool of scaling factors of (a) λ ¼0.2, (b) λ ¼0.5, and (c) λ ¼1 obtained by laser scanning microscopy.
51
the surface properties and dimensions for each scale dimension are
summarized in Table 1. The normal force was also scaled on the
basis of the thickness ratio, so that the stress state, such as contact
pressure, remained constant for different scale dimensions. The
drawing speed was 1 mm/s and the sliding length was fixed at
40 mm. The tests were conducted for three times in each scale
dimension to confirm the reproducibility of the measured data.
Fig. 4 shows the transition of the measured friction coefficient for
each specimen thickness. The representative data were chosen from
the three tests in each scale dimension. The measured friction
coefficient exhibits a clear difference for the different scale dimen-
sions. The 100 μm-thick foil has the highest friction coefficient, while
the 20 μm-thick foil has the lowest friction coefficient, with an
intermediate value for the 50 μm-thick foil. Fig. 5 summarizes the
average value of the coefficient of friction and its standard deviation
in three tests at the different scale dimensions. It can be seen that the
coefficient of friction decreases with decreasing scale dimension
under dry sliding friction [15]. To discuss the cause of this scale-
dependent behavior, a method of classifying the size effect based on
the scale-dependent model developed by Bhushan et al. [21–23] is
proposed in the following section and its applicability to describe the
tribological size effect under dry friction is investigated.
3. Theory
3.1. Scale dependent model for dry friction
According to the surface deformation and adhesion model of
friction [24], the scale dependent model developed by Bhushan
et al. [21–23] expresses the total coefficient of friction as the sum
of the adhesion component μa, deformation component μd, plow-
ing component μp, and ratchet component μr. They implemented
the scale dependence in terms of the corresponding real areas of
contact, Aa, Ad, and Ap and the corresponding shear strengths, τa,
τd, and τp as following Eq. (1);
μ ¼ μa þ μd þ μp þ μr
F a þF d þ F p
¼ þ μr
W
TD TD
RD 20μm RD 20μm
Drawing direction
Drawing direction
20μm 20μm
52 T. Shimizu et al. / Wear 330-331 (2015) 49–58

Aa U τa þ Ad U τd þ Ap U τp statistically stored dislocations (SSDs) ρS and geometrically neces-

¼ þ μr ð1Þ
W sary dislocations (GNDs) ρG as follows [25].
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
The three main scale effects under dry friction was proposed; τY ¼ τY0 1 þ ρG =ρS ¼ τY0 1 þ ld =a ð2Þ
(1) the scale effect of yield and shear strength due to strain
gradient plasticity and dislocation-assisted sliding, (2) the scale Here, τY0 is the macroscale value of the yield strength, a is the
effect of the surface geometry and contact parameters, and (3) the contact radius, and ld is the plastic deformation length, which
scale effect of three-body deformation. In the following subsec- characterizes the depth dependence of the shear yield strength.
tion, each scale dependence is briefly explained to clarify the Secondly, regarding the dislocation-assisted sliding, Hurtado
relative parameters used to classify the scale dependence. For a [26] and Bhushan et al. proposed the existence of a microslip
detailed description of the above model, see [21–23]. effect, where the sliding is assisted by dislocations [21]. On the
basis of the geometrical relation between the dislocation length ls

Table 1
Input parameters in scaled foil strip drawing test for each scale dimension.

Scaling factor (λ) 0.2 0.5 1

Foils Initial thickness (t0) [μm] 20 50 100

Yield stress (σY) [MPa] 623 610 508
Surface hardness (H) [GPa] 3.23 3.17 2.64
Young's modulus (E1) [GPa] 110 110 110
Surface roughness (Ra1) [μm] 0.07 0.05 0.12

Tools Corner radius, (rcorner) [mm] 0.95 2.38 4.75

Surface hardness, (H) [GPa] 24 24 24
Young's modulus (E2) [GPa] 750 750 750
Surface roughness (Ra2) [μm] 0.03 0.04 0.03

Applied normal force (W) [N] 10 25 50

0.35 and contact radius a, the corresponding shear strength τa during

Scaling factor λ=1 C5191-H Phosphor bronze sliding is described as
Coefficient of friction μ

0.3 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0.25 τa ¼ τa0 1 þ ls =a; ð3Þ

0.2 where τa0 is the shear strength during sliding in the limit of a⪢ls.
Thus, the ratio ls/a is the characteristic microslip parameter.
0.15 λ=0.5
0.1 Strip thickness: t =100 λ [μm]
λ=0.2 λ=0.2 Normal force: W=50 λ [N] 3.1.2. Scale effect of surface geometry and contact parameters
0.05 λ=0.5 Sliding speed: v =1[mm/s] The measured surface geometrical characteristics, such as
λ=1 Sliding distance: D =40[mm] the size and shape of the asperities, depend on the short- and
0 long-wavelength limits of the measurement. On the basis of
0 10 20 30 40
this concept, rough surface profiles are statistically character-
Sliding distance Dslide [mm] ized by a Gaussian height distribution with standard deviation
Fig. 4. Evolution of friction coefficient, μ, as a function of sliding distance, Dslide, in of the profile height σ and correlation length β n. σ is a measure
scaled foil strip drawing tests at scaling factors of λ ¼0.2, 0.5, and 1 for C5191-H of the height distribution and β n is a measure of the spatial
phosphor bronze foils. distribution. Both σ and β n initially increase with the scan size
of the measurement and approach a constant value at a certain
0.30 scan size [23]. Bhushan et al. defined this boundary scan size
C5191-H Phosphor bronze as the long wavelength limit Llwl. When the characteristic
0.25 length L of a certain contact problem is less than L lwl , scale
Coefficient of friction μ

dependence of the roughness profile parameter exists. The size

0.25 dependences of σ and β n are expressed as an empirical power
0.22 rule as follows;
 
0.19 L n
σ ¼ σ0 L o Llwl ð4Þ
0.20 Llwl
Strip thickness: t =100 λ [μm]
Normal force: W=50 λ [N]  m
L
Sliding speed: v =1[mm/s] βn ¼ βn0 L o Llwl ð5Þ
Sliding distance: D =40[mm] Llwl
0.15
0 0.2 0.4 0.6 0.8 1 1.2 where n and m are the indices of the corresponding compo-
nents and σ 0 and β0 are the macroscale values of σ and β n,
n
Scaling factor λ
respectively. Additionally, since the surface geometry varies
Fig. 5. Scale dependence of the friction coefficient under dry sliding friction in
with the surface pressure, the dependence of the limit scale
scaled foil strip drawing tests.
length Llc on the applied load is expressed as
3.1.1. Scale effect of yield and shear strength at interface !1=ðn  mÞ
βn0 pa
According to the theory of strain gradient plasticity, the shear Llc ¼ Llwl U U ; ð6Þ
yield strength τY can be written as a function of the densities of
σ 0 pac
 T. Shimizu et al. / Wear 330-331 (2015) 49–58 53

where pa is the apparent pressure and p ac is the critical probability of trapped wear particles. A square contact region with
apparent pressure below which the scale dependence occurs. length L and a density of η particles per unit area is modeled. The
Consequently, under the premise that L o Llc, the scale- distribution of the wear particle size d is given by the probability
dependent real area of elastic contact Are can be described as density function p(d). The main point of this concept is that the
 m  n wear particles in the border region with width d/2 are likely to
L βn leave the contact zone with a certain probability α, whereas the
Are ¼ Are0 p W L o Llc : ð7Þ
Llc σ En particles at the center of the contact region are likely to be trapped
and plow the contact surface.
The probability density of the size distribution of the trapped
3.1.3. Scale effect of three body deformation (plowing) particles ptr(d) is therefore given as
Although the scale-dependent model proposed above is based !  
2αðL  dÞd þ 2αd 2α d
on the long-wavelength limit Llwl of the rough surface profile, the 2
L
ptr ðdÞ ¼ pðdÞ 1  ¼ pðdÞ 1  ; do ð8Þ
concept of this scale dependence of three-body deformation, L2 L 2α
which is contributed to by wear debris and hard asperities, is
explained by the size dependence of the probability that a wear
Then the ratio of the number of trapped particles to the total
particle of a given size is trapped at the contact interface. Fig. 6
number of particles, ntr, and the average diameter of the trapped
illustrates the main concept of this size dependence of the
particles d can be calculated as

Region particles release R L=2 R L=2  

d/2 p ðdÞdd pðdÞ 1  ð2αd=LÞ dd
L ntr ¼ 0R 1 tr ¼ 0 R1 ; ð9Þ
Wear particle 0 pðdÞdd 0 ptr ðdÞdd

d Leave with a R L=2

probability α d U ptr ðdÞdd
d¼ 0
R L=2 : ð10Þ
0.5L 0 pðdÞdd

By using these parameters, the contact area of wear particles

0.2L during plowing Ap can be calculated as

2 2
d d
Ap ¼ N tr π ¼ ηL2 ntr π : ð11Þ
4 4
Region particles are trapped
Less trapped particles Less μp Thus, with decreasing scale dimension, the contact area of wear
particles decreases; thus, the friction resistance due to the plowing
Fig. 6. Schematic illustration of wear debris in the contact zone and at its border by the three-body deformation of wear particles is expected to
region [21]. decrease.

Fig. 7. Flow chart of classification of size effect of dry friction.

54 T. Shimizu et al. / Wear 330-331 (2015) 49–58
Foil strip
Compression tools
Compression tools
λ=0.2 λ=0.5 λ=1
(t0=20μm) (t0=50μm) (t0=100μm)
Fig. 8. Schematic illustrations of FE model for scaled strip drawing test at different
scale dimensions; (a) λ ¼0.2, (b) λ ¼0.5, and (c) λ ¼1.
3.2. Flow of the classification of size effect
According to the expressions for the contact parameters in the
above scale-dependent model, relevant parameters that determine
the existence of the size effect of the friction can be listed as
follows; (a) the characteristic length in the problem, which we here
define as the apparent contact length L (b) the long-wavelength
limit Llc which can be determined from the surface geometrical
factor and the surface pressure state, (c) the size of the wear
particles d and (d) the plastic deformation length, ls or ld, which is
physically related to the average distance a dislocation travels. The
geometrical scale factors for μa, μd, and μr are based on the long-
wavelength limit Llc, and for μp, the scale factors are the relative
number of trapped wear particles ntr and their mean diameter d.
Thus, the relation between L and the other scaling parameters is of
great importance in determining the existence of a size effect.
If L 4Llc and L⪢d, there is no size effect, meaning that classic
macroscopic tribology can be applied. If L oLlc and d is relatively
large compared with L, the scale-dependent model can be applied.
In addition, if L is relatively small compared with the dislocation-
related length parameter, ls or ld, the scale effects of the shear
strength and yield strength should also be considered.
On the basis of this statement, a flow chart for the classification
of the size effect under dry friction is proposed to identify the
existence of the size effect, as shown in Fig. 7. The first important
classification is to identify the long-wavelength limit for the
contact parameter Llc given by Eq. (6). The relative parameters
for this operation are the surface roughness parameters σ and βn
and the normalized apparent surface pressure pa/pac. The next step
is to determine the elastic or plastic contact regime from the
plasticity index ψ [27]. Ψ is given by the following equation,
assuming that the surface roughness have a curvature radius β.
sffiffiffiffiffi
En β assumed to be an isotropic elastoplastic body and the material
ψ¼ U ð12Þ
2H Rn
Here, En is the effective plastic modulus of the two contact
surfaces, H is the hardness of the softer material, and Rn is the
composite roughness of the two contact surfaces. In the case of
Lo Llc, the flow for the scale-dependent contribution will follow
Bhushan's scale-dependent model. The scale dependence is
divided into two cases of an increasing or decreasing adhesion
component μa. This classification depends on the plastic/elastic
contact state and ls/a, which can be described as Ls in a multiple-
asperity contact problem [22].
On the other hand, in the case of L 4Llc with plastic contact, the
next important classification is the extent of the wear behavior
and the generation of wear particles, since the plowing by wear
particles is a dominant component in this contact state. In this
classification, we adopted the following estimate for the size of the
adhesive wear particles [10];
6EW ad
dZ ð13Þ
ν2 σ y 2
where Wad is the surface energy of adhesion, ν is Poisson's ratio, E
is Young's modulus, and σY is the yield stress of the work material.
If the generated adhesive wear particles are sufficiently smaller
than the characteristic length L, the plowing size effect might be
low. Then, the final classification in this contact state is to
determine the occurrence of abrasive wear. Kayaba et al. proposed
the following formula for the penetration depth, Dp [28];
  !1=2
π H 1=2 β2 π H
Dp ¼ β  1 : ð14Þ
2W 2W
When Dp 40.2, the abrasive wear regime becomes cutting-type
wear and abrasive wear particles are generated.
In the case of elastic contact or no effect of plowing by the wear
particles, this contact state might be classified as having no size
effect as shown in Fig. 7. By using this flow chart, the occurrence of
the size effect under dry friction can be classified. In the following
section, the practical problem of the scaled strip drawing test
discussed in Section 2 is analyzed by this classification method.
4. Theoretical analysis
4.1. Characteristic length
Coming back to the experimental observation discussed in
Section 2, the cause of the tendency of the coefficient of friction
to decrease under dry sliding contact in the scaled foil strip
drawing test is analyzed using the flow chart shown in Fig. 7.
As mentioned in the previous section, the identification of the
characteristic length L is of great importance in discussing the
scale dependence of dry friction. In the strip drawing experiment,
the contact length in the drawing direction was defined as the
characteristic length. Since the drawing tool geometry includes
circular arcs, it is difficult to measure the actual contact length
during the strip drawing test. To identify the contact length of each
scaled dimension in the strip drawing test, a finite element (FE)
analysis was carried out. A simulation was performed using an
explicit dynamic FE code (LS-DYNA ver.971). Schematic illustra-
tions of the model at each scale dimension are shown in Fig. 8.
For each thickness, a sliding length of 2 mm was chosen as the
modeling region. A hexahedral 8-node mesh was used, and only
1 element was meshed to the width direction, assuming a plane
strain state for the friction test. Mesh size of 2 μm  2 μm  2 μm
was used for each foil thickness. Numbers of elements were 7500
for λ ¼0.2, 19500 for λ ¼0.5 and 31,500 for λ ¼1.0. The foil was
properties were modeled using the flow curves obtained by tensile
tests of the phosphor bronze foils [29]. The compression tools
were assumed to be rigid bodies. As a contact condition, the
friction force was calculated by using a node constraint method
with Coulomb's law. The friction coefficients between the blank
and the tools were adjusted by comparing the output analytical
data of tensile force with the obtained experimental data. The
conditions of the FE analysis are summarized in Table 2.
Fig. 9 shows example data of the variation of the contact length
during the sliding of the tools for the 20 μm-thick foil. The contact
length first increases at the compression stage, and then after the
drawing has started, the contact length approaches a constant
value during the sliding. This constant contact length was defined
as the characteristic length for this contact problem. Similarly, the
 T. Shimizu et al. / Wear 330-331 (2015) 49–58 55

contact lengths for the 50- and 100 μm-thick foils were calculated, 40
as summarized in Fig. 10. A linear relation between the scaling Compression Scaling factor λ=0.2
35
factor λ and contact length L was successfully obtained as required stage
for experiments to investigate the size effect. On the basis of these 30

Contact length L [μm]

obtained contact lengths, the applied apparent pressure in the Drawing stage
25
entire range of scale dimension was found to be approximately
420 MPa. 20 Constant contact length L:12μm
15
4.2. Long-wavelength limit
10
According to the proposed flow chart, the first classification is 5
the identification of the long-wavelength limit Llwl. To identify Llwl,
0
the scan size dependences of σ and βn were analyzed from the 0.0 0.2 0.4 0.6 0.8 1.0
actual surface data of a foil strip and tool surface. Assuming that Time step [ms]
the statistical characteristic of each scale dimension has a similar
tendency, the surface profiles of the foil strip and the tool at the Fig. 9. Example of FE analytical results of variation of contact length L during strip
drawing test at scale dimension of λ¼ 0.2.
scaling factor of λ ¼0.2, as shown in Fig. 11, were analyzed to
obtain a representative value. To characterize the profiles of the
surface in contact, composite roughness parameters, given by the
following equations, are calculated.
60 FEM results of 54
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
contact length L
σ ¼ σ 1 2 þ σ 22 ð15Þ

Contact length L /μm

50
R² = 0.99
1 1 1
n ¼ nþ ð16Þ 40
β β1 βn2 31
30
σ and βn for each strip and tool surface were calculated with
the scan size ranging from 1 to 70 μm. For the calculations of βn,
20
autocorrelation functions were introduced. Fig. 12 shows example 12
data of the autocorrelation function for the surface profile of the 10 Strip thickness: t0=100㽢λ [μm]
20 μm-thick strip specimen as a function of the autocorrelation Normal force: W=50㽢λ [N]
interval k, expressed as the distance between two analyzed points. 0
0 0.2 0.4 0.6 0.8 1 1.2
Here, βn is defined as the distance k at which the autocorrelation
function first drops to the small fraction near to zero. This suggests Scaling factor λ
that the distance between the two analyzed points, which exceeds Fig. 10. FE analytical data of the variation of contact length L as a function of scaling
the βn value, is statistically independent. By performing the factor λ.
analysis at different scan sizes, the scale dependence of βn was
plotted.
Fig. 13 shows the calculation results of σ and βn at the scaling 0.15
factor of λ ¼0.2. Both σ and βn tend to decrease with decreasing Phosphor Scaling factor λ=0.2
Roughness height hr /μm

bronze foil
scan size. Note that the long-wavelength limit Llwl for σ and βn is 0.1
between approximately 40 and 45 μm. In this calculation, the
average value of 42 μm is used for Llwl. The values of the indices of 0.05
m and n for Eqs. (4) and (5) were found to be m ¼0.5 and n ¼ 0.2 by
0
curve fitting. Additionally, the macroscale values of σ0 and β0 were
n

defined as σ0 ¼ 46 nm and β 0 ¼ 1:26 μm.

n
-0.05
Furthermore, to define Llc for this contact pair, the dependence
of the critical apparent pressure pac on Llc was evaluated. Fig. 14 -0.1 Foils
shows the relation between pac and the normalized long- Tools Compression tool
wavelength limit Llc/Llwl under several apparent pressures. It is -0.15
0 20 40 60
clearly demonstrated that Llc/Llwl increases with increasing pac.
However, Llc/Llwl decreases with increasing applied apparent pres- Scanned position xscan /μm
sure. For the applied apparent pressure of 420 MPa in the scaled Fig. 11. Surface roughness profiles of drawing strip specimen and compression tool
strip drawing experiment, which was obtained by FEM analysis, for scaling factor of λ¼ 0.2 obtained by confocal laser scanning microscopy.

Table 2
Input mechanical properties of phosphor bronze foils and conditions of FE analysis.

Scaling factor, (λ) 0.2 0.5 1 Drawing tool

Strip thickness (t0) [μm] 20 50 100

Material model Elastic–plastic body Rigid body

Mass density (ρ) [mg/μm  3] 8.83  10  7 1.42  10  6
Young’s modulus (E) [GPa] 110 630
Yield stress (σY) [MPa] 622 610 508 –
Poisson's ratio (ν) 0.3 –
Static/kinetic friction coefficient (μs, μk) μs ¼ 0.16, μk ¼0.14 –
56 T. Shimizu et al. / Wear 330-331 (2015) 49–58

phosphor bronze : 3.6GPa

Hardness of
Normalized long wavelength limit
1.4
Phosphor bronze t0=20μm
Autocorrelation funcrtion rk
0.8 1.2
Llwl
0.6 1
Correlation length β*
0.8

Llc/Llwl
0.4
0.6
0.2
0.4
0
0.2
-0.2
0
Scan size: 40μm 0 2 4 6 8 10
-0.4
k=1.25 Critical apparent pressure pac /GPa
0 2.25 4.5 6.75 9 11.25
Autocorrelation interval k Fig. 14. Variations of normalized long-wavelength limit Llc/Llwl as a function of
critical apparent pressure pac, under several apparent pressure conditions.
expressed as a distance [μm]
Fig. 12. Example data of autocorrelation analysis for surface profile of 20 μm-thick
phosphor bronze foil specimen (scan size: 20 μm).
2
Phosphor bronze foil vs. WC-Co tool 1.77
1.8
160 Phosphor bronze foil vs. WC-Co tool
1.6
1.38

Plasticity index ψ
140
σ /nm and β* / 10-2μm

1.4

contact
Plastic
120 1.2

Transition
100 β*0 1

regime
80 β* 0.8 0.67
σ0 σ
0.6

Elastic
contact
60
0.4
40 0.2
20 L lwl 0
0.2 0.5 1
0
0 10 20 30 40 50 60 70 Scaling factor λ
Scan size Lscan /μm Fig. 15. Comparison of plasticity index between the contact pair at different scale
dimensions in the foil strip drawing test.
Fig. 13. Variations of statistical surface roughness parameters σ and β* as a function
of scan size for a phosphor bronze rolled foil in contact with a WC–Co
compression tool. 4.4. Three-body plowing

Llc/Llwl is less than 0.6, even for the impractical pac value of 10 GPa. To determine the probability of the occurrence of the size effect
Since pac is relevant to the ratio of real contact area to apparent due to three-body deformation, the wear particle size was first
contact area, it is strongly related to the softer material property of estimated for each scale dimension. Table 3 summarizes the input
the contact pair. If we assume pac to be the hardness of the parameters and calculated wear particle sizes. Although the
phosphor bronze foil, H¼3.6 GPa, then Llc/Llwl is close to zero under calculated values are slightly different for each scale due to the
an apparent pressure of 420 MPa. This suggests that Llc is consider- difference in the yield stress for each thickness, the particle size
ably low, meaning that it might not be in the range of the scale appears to be comparable to the calculated contact length as
dependence due to the long-wavelength limit under this contact shown in Fig. 9. Thus, the scale-dependent model of three-body
state. In contrast, at the lower apparent pressure of 100 MPa, Llc/Llwl deformation appears to be applicable to the contact state in foil
becomes more than 1, under the assumption of pac ¼H. Since the strip drawing tests.
contact length L appears to decrease at a lower applied pressure, the
probability that the scale effect exists owing to the long-wavelength 4.5. Total friction coefficient
limit might increase.
Using the classification process in the previous section, it was
found that the scale effect of friction due to the long-wavelength
limit appears to be relatively low and that the cause of the
4.3. Plasticity index decreasing tendency of the coefficient of friction might be the
scale effect of three-body deformation induced in the contact state
To evaluate of the plastic/elastic contact state at each scale in the foil strip drawing tests. Thus, the components of the
dimension, the plasticity index ψ as given by Eq. (12) was coefficient of friction for μa, μd, and μr can be calculated assuming
calculated. The input variance for each scale dimension was a macroscale tribology. Using the results of classical tribology,
determined from the practical surface properties given in these three components were calculated using following Eqs. (
Table 1. Fig. 15 shows the results of the calculation. For the scale 17)–(19) [10,24]. The input parameters are summarized in Table 4.
dimension of λ ¼0.2, ψ is higher than 0.6, but less than 1, which !
can be identified as the transition from the elastic to plastic τ 1
μa ¼ a   ð17Þ
contact regime. In contrast, for λ ¼ 0.5 and 1, ψ exceeds 1, which H 1 2W ad = rH sin θ
suggests that the contact state is plastic contact. Thus, since
ψ 40.6 was obtained in the entire range of scale dimension, scale 2 tan θ
μd ¼ ð18Þ
effect of three-body plowing should be take into consideration. π
 T. Shimizu et al. / Wear 330-331 (2015) 49–58 57

Table 3 Table 5
Input parameters and calculated results of adhesive wear particle size for each Parameters used in the calculations of μp components of the coefficients of friction
scaling factor. in scaled strip drawing test.

Scaling factor (λ) 0.2 0.5 1 Scaling factor (λ) 0.2 0.5 1

Young's modulus (E) [GPa] 110 110 110 Contact length (L) [μm] 12 31 54
Yield stress (σY) [MPa] 622 610 508 Applied load (W) [N] 10 25 50
Poisson's ratio (ν) 0.38 0.38 0.38 Strip yield shear strength (τY) [MPa] 435 427 356
Energy of adhesion (Wad) [J/mm2] 2.7  10  5 2.7  10  5 2.7  10  5 Standard deviation of wear particle size (σ) [μm] 0.16 0.17 0.24
Adhesive wear particle size [μm] 0.32 0.34 0.48 Probability of leaving particle (α) 0.9 0.9 0.9
Particle density (η) [particles/mm2] 100 100 100

100
Table 4 99.5 99.6
ntr

Real area of contact Ap [μm2]

Parameters used in the calculations of the μa, μd, and μr components of coefficient of 99.5
10000
friction in the scaled strip drawing test. 98.8 13220
99

Scaling factor (λ) 0.2 0.5 1 98.5

98
Applied load (W) [N] 10 25 50 2090
1000
Strip yield shear strength (τY) [MPa] 435 427 356 97.5 Ap
Strip hardness (H) [GPa] 3.23 3.17 2.64 97
Energy of adhesion (Wad) [J/mm2] 2.7  10  5 2.7  10  5 2.7  10  5 Ap
96.5 283 ntr
Mean junction radius (rmean) [μm] 7.16 2.78 3.80
Tool roughness angle (θ) [deg] 0.41 2.03 1.46 96 100
0 0.2 0.4 0.6 0.8 1 1.2
Scaling factor λ

μr ¼ μa tan θ 2
ð19Þ Fig. 16. Variations of the ratio of trapped particles ntr and real area of contact Ap as
a function of scaling factor λ.
For the calculation of μp, the scale-dep.endent model described
by Eq. (11) was applied. In the calculation of the probability
0.3
density function of the particle size given by Eq. (8), an exponen- Adhesion comp. μa Phosphor bronze C5191-H
tial distribution was assumed. By substituting the exponential Deformation comp. μd
Total friction coefficient μtotal

0.25 Plowing comp. μp

probability density function into Eqs. (9) and (10), the following Ratchet comp. μr
equations can be derived [22]. Experimental result μr
   
2ασ L 0.2
ntr ¼ exp  1 þ1 ð20Þ
L 2ασ μd μ p

    0.15
exp  ðL=2ασ Þ U 1 þ ð4ασ =LÞ þð1 ð4ασ =LÞÞ
d ¼σU    
ðð2ασ =LÞ exp  ðL=2ασ Þ  1 þ1Þ
Here, we assume the probability to be α ¼0.9, which means that
90% of the particles at the border region will leave the contact
surface. In addition, the standard deviation of the particle size was
defined as 50% of the wear particle size given in Table 3. Moreover,
we assumed that 100 particles exist per 1 mm2 for each scale
dimension to define the density of the particles η. This value was
roughly estimated from the number of plowing scratches observed
on the drawn strip samples after the strip drawing tests [15]. By
substituting the relevant value into following Eq. (22), the coeffi-
cient of friction for the plowing component was calculated for
each scale dimension.
2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Ap τ p L2 d
μp ¼ ¼ ¼ η U π U ntr U τY0 1 þ 2ld =d
W W 4 geometrical/mechanical properties, the scale effect of dry friction
The input variances used in this calculation are summarized in
Table 5. Fig. 16 shows the calculation results of the ratio of trapped
particles ntr and the real area of contact Ap, for each scale dimension.
The decreasing tendency of the ratio of trapped particles strongly
affects the decrease in real area of contact, which might affect the
decreasing tendency of the plowing components with decreasing λ.
Fig. 17 summarizes the analyzed total friction coefficient com-
pared with the experimental data obtained in Section 2. Although
the analytical results slightly underestimate the experimental results,
it is clearly demonstrated that the difference in the friction coefficient
between different scale dimensions is mainly due to the plowing
component. Since the relative ratio of the number of loosened wear
particles should increase with decreasing scale dimension, the
friction coefficient appears to decrease with downscaling.
Experimental result
ð21Þ
0.1 μa
0.05
0
λ=0.2 λ=0.5 λ=1
Scaling factor λ
Fig. 17. Analyzed total friction coefficient along with its contributions of the four
components at different scaling factors, compared with experimental data.
Consequently, by following the proposed classification app
ð22Þ roach based on the analysis of the contact state and surface
observed in the foil strip drawing test was found to be classified as
the size effect of the plowing components. Thus, in the contact
state of dry sliding in microforming operation, the dominant
tirbological factor appears to be the contribution of the plowing
resistance by the generated wear particles. In fact, in a scaled
progressive dry micro-deep drawing test of stainless steel, it was
demonstrated that the process with a larger scale in the mm range
resulted in strong adhesion of the work materials on the forming
tool surface after 300 drawing tests, while the variation of the tool
surface was relatively low for the μm-scale process [30]. This
suggests that the accumulation of wear particles is much lower at
micro-scale dimensions, since the probability of the ejection of
wear particles from the contact surface is higher. Therefore,
surface design to reduce the generation of wear particles might
58 T. Shimizu et al. / Wear 330-331 (2015) 49–58
be effective for improving of the mesoscopic tribological proper-
ties and their stability in lubricant-free microforming operation.
5. Summary and conclusions
The existence of the size effect under dry sliding friction strongly
depends on the contact state and the geometrical and mechanical
properties of the surface. To generally describe the size effect under
dry friction by a coherent analytical method, we proposed a
classification approach in which the relations between character-
istic length L in the contact problem and three significant criteria;
(1) the long wavelength limit Llc (2) the size of the wear particles d,
and (3) the plastic deformation length ls or ld, were compared. By
applying this approach to analyze the decreasing tendency of the
friction coefficient in dry sliding friction in the foil strip drawing
test, its applicability and the positioning in the transitional scale
range of mesoscopic tribology in microforming operation were
quantitatively analyzed. The following conclusions were drawn.
 It was found that the size effect due to the geometrical para-
meters of the surface and the yield and shear strength is small
owing to the relatively large contact length L compared with the
calculated Llc. However, the calculation results for adhesive wear
particles suggested that the contact state in foil strip drawing
might be in the range of the size effect of the plowing component.
 By estimating the four components of the friction coefficient, it was
demonstrated that the decreasing tendency of the total friction
coefficient in the foil strip drawing test was mainly due to the size
effect of the plowing component. This, in turn, suggests that the
control of wear particle generation plays an important role in
improving the mesoscopic tribology under dry sliding friction.
Consequently, the applicability of the proposed classification
approach was successfully demonstrated and the positioning of the
contact state in the micro-sheet-metal forming was clarified. The
characterization presented in this study is only an example and does
not cover all contact states in microforming. Under a relatively low
contact pressure of less than 100 MPa with a shorter contact length
than that for the contact state demonstrated in the present study,
another type of size effect due to the long-wavelength limit and the
yield and shear strengths might become effective. In contrast, under
a higher contact state of more than 1 GPa, such as the typical contact
state in micro-bulk forming, the size effects stated in the proposed
model might completely disappear owing to the large characteristic
length for a plastic contact state, as demonstrated in previous reports
[6,7]. Thus, conventional macroscopic tribology can be applied to the
surface design of tools and work materials in such a contact state.
To investigate the applicability of the proposed method in a
more practical contact state such as that in micro-sheet-metal
forming, we will carry out quantitative analysis of the contact state
in the micro-deep drawing process as a future task.
Acknowledgments
The authors gratefully acknowledge funding from Japan Society for
the Promotion of Science (JSPS), in the form of a Grant-in-Aid for Young
Scientists B (No. 26820327), and from the AMADA foundation for a
Grant-in Aid for general research and development (AF-2013028).
Furthermore, the authors would like to express their gratitude to Mr.
Kuniyoshi Ito of Micro Fabrication Laboratory LLC. for supporting the
design and preparation of the foil strip drawing equipment.
References
[1] M. Geiger, M. Kleiner, R. Eckstein, N. Tiesler, U. Engel, Microforming, CIRP Ann.
50–2 (2001) 445–462.
[2] F. Vollertsen, D. Biermann, H.N. Hansen, I.S. Jawahir, K. Kuzman, Size effect in
manufacturing of metallic components, CIRP Ann. 58 (2009) 566–587.
[3] N. Tiesler, Microforming – size effects in friction and their influence on
extrusion processes, Wire 52 (2002) 34–38.
[4] F. Vollertsen, Z. Hu, Tribological size effects in sheet metal forming measured
by a strip drawing test, CIRP Ann. 55-1 (2006) 291–294.
[5] U Engel, Tribology in microforming, Wear 260 (2006) 265–273.
[6] L. Mori, N. Krishnan, J. Cao, H. Espinosa, Study of the size effects and friction
conditions in micro-extrusion. Part II. Size effect in dynamic friction for brass–
steel pairs, ASME J. Manuf. Sci. Eng. 129-4 (2007) 677–689.
[7] A. Messner, U Engel, R. Kals, F. Vollertsen, Size effect in the FE-simulation of
micro-forming processes, J. Mater. Proces. Technol. 45 (1–4) (1994) 371–376.
[8] M. Geiger, A. Messner, U. Engel, Production of microparts – size effects in bulk
metal forming, similarity theory, Prod. Eng. 4-1 (1997) 55–58.
[9] B. Guo, F. Gong, C.J. Wang, D.B. Shan, Flow stress and tribology size effects in
scaled down cylinder compression, Trans. Nonferrous Met. Soc. China 19
(2009) 516–520.
[10] E. Rabinowicz, Friction and Wear of Materials, Wiley, New York, 1995.
[11] N.P. Suh, H.C. Sin, The genesis of friction, Wear 69 (1981) 91–114.
[12] W. Zheng, G. Wang, G. Zhao, D. Wei, Z. Jiang, Modeling and analysis of dry
friction in micro-forming of metals, Tribol. Int. 57 (2013) 202–209.
[13] T. Shimizu, H. Komiya, T. Watanabe, Y. Teranishi, H. Nagasaka, K. Morikawa,
M. Yang, HIPIMS deposition of TiAlN films on inner wall of micro-dies and its
applicability in micro-sheet metal forming, Surf. Coat. Technol. 250 (2014) 44–51.
[14] T. Shimizu, K. Manabe, M. Yang, Deformation behavior of ultra-thin metal foils
in strip drawing friction test, Key Eng. Mater. 443 (2010) 110–115.
[15] T. Shimizu, S. Iwaoka, M. Yang, K. Manabe, Scale dependence of dry friction in
micro sheet metal forming (ICTP2011), Steel Res. Int.- Spec. Ed. (2011) 979–984.
[16] O. Guillon, X. Roizard, P. Belliard, Experimental methodology to study
tribological aspects of deep drawing – application to aluminum alloy sheets
and tool coatings, Tribol. Int. 34 (2001) 757–766.
[17] R. Ahmed, M.P.F. Sutcliffe, An experimental investigation of surface pit
evolution during cold-rolling or drawing of stainless steel strip, ASME J. Tribol.
123 (2001) 1–7.
[18] M.P.F. Sutcliffe, H.R. Le, D. Farrugia, Simulation of transfer layer formation in
strip drawing of stainless steel, Wear 254 (2003) 523–531.
[19] H.L. Costa, I.M. Hutchings, Effect of die surface pattering on lubrication in strip
drawing, J. Mater. Process. Technol. 209 (2009) 1175–1180.
[20] V. Servo, L Vilhena, P.N. Dias, D. Becker, S. Wagner, A. Cavaleiro, Tribological
behavior of W–Ti–N coatings in semi-industrial strip drawing test, J. Mater.
Process. Technol. 209 (2009) 4662–4667.
[21] B. Bhushan, M. Nosonovsky, Scale effects in friction using strain gradient plasticity
and dislocation assisted sliding (Microslip), Acta Mater. 51 (2003) 4331–4345.
[22] B. Bhushan, M. Nosonovsky, Comprehensive model for scale effects in friction
due to adhesion and two- and three- body deformation (plowing), Acta Mater.
52 (2004) 2461–2474.
[23] M. Nosonovsky, B. Bhushan, Scale effect in dry friction during multiple-
asperity contact, ASME J. Tribol. 127 (2005) 37–46.
[24] B. Bhushan, Introduction to Tribology, Wiley, New York, 2002.
[25] W.D. Nix, H. Gao, Indentation size effects in crystalline materials A law for
strain gradient plasticity, J. Mech. Phys. Sol. 46 (3) (1997) 411–425.
[26] J. Hurtado, K.S. Kim, Scale effect in friction of single-asperity contacts. I. From
concurrent slip to single-dislocation-assisted slip. II. Multiple-dislocations-
cooperated slip, Proc. R. Soc. Lond. Ser. A 455 (1999) 3363–3400.
[27] J.A. Greenwood, J.B.P. Williamson, Contact of nominally flat surfaces, Proc. R.
Soc. Lond, Ser. A 295 (1966) 300–319.
[28] T. Kayaba, K. Hokkirigawa, K. Kato, Analysis of the wear mechanism by
successive observation of wear processes in a scanning electron microscope,
Wear 110 (1986) 419–430.
[29] T. Shimizu, M. Ogawa, M. Yang, K. Manabe, Plastic anisotropy of ultra-thin
rolled phosphor bronze foils and its thickness strain evolution in micro-deep
drawing, Mater. Des. 56 (2014) 604–612.
[30] T. Shimizu, Y. Murashige, S. Iwaoka, M. Yang, K. Manabe, Scale dependence of
adhesion behavior under dry friction in progressive micro-deep drawing,
J. Sol. Mech. Mater. Eng. 7 (2013) 251–263.
Another random document with
no related content on Scribd:
 INTRODUCTORY ESSAY.
BY J. ARTHUR THOMSON.

BREHM’S PLACE AMONG NATURALIST-TRAVELLERS.

Though Brehm’s lectures might well be left, as his son has said, to
speak for themselves, it seems useful to introduce them in their
English dress with some notes on the evolution of the naturalist-
traveller and on Brehm’s place in the honourable list; for an
adequate appreciation of a book like this depends in part on a
recognition of the position it occupies among analogous works, and
on having some picture of the illustrious author himself.
In sketching the history of the naturalist-traveller it is not necessary
to go very far back; for though it is interesting to recall how men of
old followed their migrating herds, as the Lapp or Ostiak does his
reindeer, and were led by them to fresh fields and new conquests, or
how others followed the salmon down the rivers and became the
toilers of the sea, this ancient lore is full of uncertainty, and is,
besides, of more moment to the sociologist than to the naturalist.
What we attempt here is merely to indicate the various types of
naturalist-traveller who have in the course of time succeeded one
another in the quest for the new.

I.
The foundations of zoology were laid by Aristotle some three
hundred years before Christ, but they remained unbuilt on for nearly
eighteen centuries. Here and there some enthusiast strove unaided,
but only a fragmentary superstructure was reared. In fact, men were
pre-occupied with tasks of civilization more serious than the
prosecution of zoology, though that is not trivial. Gradually, however,
great social movements, such as the Crusades and the collapse of
Feudalism; great intellectual and emotional movements, such as
those of the Renaissance; great inventions, such as that of printing,
gave new life to Europe, and zoology shared in the re-awakening.
Yet the natural history of the Middle Ages was in great part mystical;
fancy and superstition ran riot along paths where science afterwards
established order, and, for all practical purposes, the history of
zoology, apart from the efforts of a few pioneers, may be said to date
from the sixteenth century.
Now, one indubitable factor in the scientific renaissance of the
sixteenth century was the enthusiasm of the early travellers, and this
stimulus, periodically recurrent, has never failed to have a similar
effect—of giving new life to science. But while science, and zoology
as a branch of it, has been evolving during the last three centuries,
the traveller, too, has shared in the evolution. It is this which we wish
to trace.
I. The Romantic Type. Many of the old travellers, from Herodotus
onwards, were observant and enthusiastic; most were credulous and
garrulous. In days when the critical spirit was young, and verification
hardly possible, there could not but be a strong temptation to tell
extraordinary “travellers’ tales”. And they did. Nor need we scoff at
them loudly, for the type dies hard; every year such tales are told.
Oderico de Pordenone and other mediæval travellers who give some
substance to the mythical Sir John de Maundeville were travellers of
this genial type. Oderico describes an interesting connecting link
between the animal and vegetable kingdom, a literal “zoophyte”, the
“vegetable lamb”, which seems to have been a woolly Scythian fern,
with its counterpart in the large fungus which colonials sometimes
speak of as the “vegetable sheep”. As for the pretended Sir John, he
had in his power of swallowing marvels a gape hardly less than that
of the great snakes which he describes. But even now do we not see
his snakes in at least the picture-books on which innocent youth is
nurtured? The basilisk (one of the most harmless of lizards) “sleyeth
men beholding it”; the “cocodrilles also sley men”—they do indeed
—“and eate them weeping, and they have no tongue”. “The griffin of
Bactria hath a body greater than eight lyons and stall worthier than a
hundred egles, for certainly he will beare to his nest flying, a horse
and a man upon his back.” He was not readily daunted, Sir John, for
when they told him of the lamb-tree which bears lambs in its pods,
his British pluck did not desert him, and he gave answer that he
“held it for no marvayle, for in his country are trees which bear fruit
which become birds flying, and they are good to eate, and that that
falleth on the water, liveth, and that that falleth on earth, dyeth; and
they marvailed much thereat”. The tale of the barnacle-tree was a
trump card in those days!
Another example of this type, but rising distinctly above it in
trustworthiness, was the Venetian Marco Polo, who in the thirteenth
century explored Asia from the Black Sea to Pekin, from the Altai to
Sumatra, and doubtless saw much, though not quite so much as he
describes. He will correct the fables of his predecessors, he tells us,
demonstrating gravely that the unicorn or rhinoceros does not allow
himself to be captured by a gentle maiden, but he proceeds to
describe tailed men, yea, headless men, without, so far as can be
seen, any touch of sarcasm. Of how many marvels, from porcupines
throwing off their spines and snakes with clawed fore-feet, to the
great Rukh, which could bear not merely a poor Sinbad but an
elephant through the air, is it not written in the books of Ser Marco
Polo of Venezia?
II. The Encyclopædist Type.—This unwieldy title, suggestive of an
omnivorous hunger for knowledge, is conveniently, as well as
technically, descriptive of a type of naturalist characteristic of the
early years of the scientific renaissance. Edward Wotton (d. 1555),
the Swiss Gesner (d. 1565), the Italian Aldrovandi (d. 1605), the
Scotsman Johnson (d. 1675), are good examples. These
encyclopædists were at least impressed with the necessity of getting
close to the facts of nature, of observing for themselves, and we
cannot blame them much if their critical faculties were dulled by the
strength of their enthusiasm. They could not all at once forget the
mediæval dreams, nor did they make any strenuous effort to
rationalize the materials which they so industriously gathered. They
harvested but did not thrash. Ostrich-like, their appetite was greater
than their power of digesting. A hasty judgment might call them mere
compilers, for they gathered all possible information from all sources,
but, on closer acquaintance, the encyclopædists grow upon one.
Their industry was astounding, their ambition lofty; and they
prepared the way for men like Ray and Linnæus, in whom was the
genius of order.
Associated with this period there were many naturalist-travellers,
most of whom are hardly now remembered, save perhaps when we
repeat the name of some plant or animal which commemorates its
discoverer. José d’Acosta (d. 1600), a missionary in Peru, described
some of the gigantic fossils of South America; Francesco Hernanded
published about 1615 a book on the natural history of Mexico with
1200 illustrations; Marcgrav and Piso explored Brazil; Jacob Bontius,
the East Indies; Prosper Alpinus, Egypt; Belon, the Mediterranean
region; and there were many others. But it is useless to multiply what
must here remain mere citations of names. The point is simply this,
that, associated with the marvellous accumulative industry of the
encyclopædists and with the renaissance of zoology in the sixteenth
and seventeenth centuries, there were numerous naturalist-travellers
who described what they saw, and not what they fancied might be
seen.
III. The General Naturalist Type.—As Ray (d. 1705) and Linnæus
(d. 1778) began to reduce to order the accumulations of the
encyclopædists, and as the anatomists and physiologists began the
precise study of structure and function, the naturalist-travellers
became more definite in their aims and more accurate in their
observations. Linnæus himself sent several of his pupils on precisely
scientific journeys. Moreover, in the eighteenth century there were
not a few expeditions of geographical and physical purpose which
occasionally condescended to take a zoologist on board. Thus
Captain Cook was accompanied on his first voyage (1768-1781) by
Banks and Solander, and on his second voyage by the Forsters,
father and son. On his third voyage he expressly forbade the
intrusion of any naturalist, but from all that we can gather it would
have been better for himself if he had not done so. In these
combined voyages there was nascent the idea of co-operative
expeditions, of which the greatest has been that of the Challenger.
In illustration of travellers who were not specialists, but in varying
degrees widely interested naturalists, it will be sufficient to cite three
names—Thomas Pennant, Peter Pallas, and, greatest of all,
Alexander von Humboldt.
Of Thomas Pennant (1726-1798) we may note that he was one of
the early travellers in Scotland, which was then, as he says, almost
as unknown as Kamchatka, and that he extorted from Dr. Johnson
the admission, “He’s a Whig, sir, a sad dog; but he’s the best
traveller I ever read; he observes more things than any one else
does”. He knew Buffon and corresponded with Linnæus, and was
the author of several works on British and North American zoology.
His so-called Arctic Zoology is mainly a sketch of the fauna in the
northern regions of North America, begun “when the empire of Great
Britain was entire, and possessed the northern part of the New
World with envied splendour”. His perspective is excellent! the
botanist, the fossilist, the historian, the geographer must, he says,
accompany him on his zoological tours, “to trace the gradual
increase of the animal world from the scanty pittance given to the
rocks of Spitzbergen to the swarms of beings which enliven the
vegetating plains of Senegal; to point out the causes of the local
niggardness of certain places, and the prodigious plenty in others”. It
was about the same time (1777) that E. A. W. Zimmermann,
Professor of Mathematics at Brunswick, published a quarto in Latin,
entitled Specimen Zoologiæ Geographicæ Quadrupedum, “with a
most curious map”, says Pennant, “in which is given the name of
every animal in its proper climate, so that a view of the whole
quadruped creation is placed before one’s eyes, in a manner
perfectly new and instructive”. It was wonderful then, but the map in
question looks commonplace enough nowadays.
Peter Simon Pallas (1741-1811) was a student of medicine and
natural science, and did good work as a systematic and anatomical
zoologist. He was the first, we believe, to express the relationships of
animals in a genealogical tree, but his interest for us here lies in his
zoological exploration of Russia and Siberia, the results of which are
embodied in a series of bulky volumes, admirable in their careful
thoroughness. We rank him rather as one of the forerunners of
Humboldt than as a zoologist, for his services to ethnology and
geology were of great importance. He pondered over the results of
his explorations, and many of his questionings in regard to
geographical distribution, the influence of climate, the variation of
animals, and similar problems, were prophetic of the light which was
soon to dawn on biological science.
Alexander von Humboldt (1769-1859) was undoubtedly one of the
greatest naturalists of the century which his life well nigh covered.
Geologist, botanist, zoologist, and more, he was almost the last of
the all-round naturalists. In this indeed lay his weakness as well as
his strength, for great breadth of view is apt to imply a lack of
precision as to details. In boyhood, “when life”, as he says, “appears
an unlimited horizon”, he had strong desires after travel, which were
in part gratified by excursions with George Forster and by Swiss
explorations with the sagacious old geographer Leopold von Buch.
These, however, only whetted his enthusiasm for journeys with a
larger radius. At length, after many discouragements, he sailed in
1799 from Corunna, with Aimé Bonpland as companion, and spent
five years in exploring the equinoctial regions of the New World. The
full record of his voyage one cannot be expected to read, for there
are about thirty volumes of it in the complete edition, but what we
should all know is Humboldt’s Personal Narrative, in which the chief
results of his explorations are charmingly set forth. Later in life
(1829) he went with Ehrenberg and Rose to North Asia, and his
crowning work was the publication of Cosmos (1845-58), which
originated in a series of lectures delivered in the University of Berlin.
In front of that building his statue now stands, along with that of his
not less famous brother Wilhelm.
We think of Humboldt not so much as an early explorer of tropical
America, nor because he described the habits of the condor and
made observations on electric eels, nor because he furnished Cuvier
and Latreille with many new specimens, but rather as a magnificent
type of the naturalist-traveller, observant, widely interested, and
thoughtful, who pointed forward to Darwin in the success with which
he realized the complexity of inter-relations in nature. Many a
traveller, even among his contemporaries, discovered more new
plants and animals than the author of Cosmos, but none approached
him as an all-round naturalist, able to look out on all orders of facts
with keenly intelligent eyes, a man, moreover, in whom devotion to
science never dulled poetic feeling. His work is of real importance in
the history of geographical distribution, for he endeavoured to
interpret the peculiarities of the various faunas in connection with the
peculiar environment of the different regions—a consideration which
is at least an element in the solution of some of the problems of
distribution. It is especially important in regard to plants, and one
may perhaps say that Humboldt, by his vivid pictures of the
vegetable “physiognomy” of different regions, and by his
observations on the relations between climate and flora, laid the
foundations of the scientific study of the geographical distribution of
plants. We find in some of his Charakterbilder, for example in his
Views of Nature, the prototype of those synthetic pictures which give
Brehm’s popular lectures their peculiar interest and value.
IV. The Specialist Type.—It would say little for scientific discipline if
it were true that a man learned, let us say, in zoology, could spend
years in a new country without having something fresh to tell us
about matters outside of his specialism—the rocks, the plants, and
the people. But it is not true. There have been few great travellers
who have been narrow specialists, and one might find more than one
case of a naturalist starting on his travels as a zoologist and
returning an anthropologist as well. Yet it is evident enough that few
men can be master of more than one craft. There have been few
travellers like Humboldt, few records like Darwin’s Voyage of the
Beagle (1831-6). Hence we recognize more and more as we
approach our own day that naturalist-travellers have been successful
either as specialists, or, on the other hand, in so far as they have
furnished material for generalization (Type V.). The specialism may
of course take various forms: a journey may be undertaken by one
who is purely an ornithologist, or it may be undertaken with one
particular problem in view, or it may be organized, like the
Challenger expedition, with the co-operation of a number of
specialists.
The French took the lead in organizing zoological expeditions. As
early as 1800 they sent out the Géographe, Naturaliste, and
Casuarina, zoologically conducted by Bury de St. Vincent, Péron,
and Lesueur. Further expeditions followed with Quoy and Gaimard,
Lesson, Eydoux, Souleyet, Dupetit-Thouars, and others as
zoological guides. The English whaling industry gave early
opportunity to not a few naturalists; and it is now a long time since
Hooker went with Sir James Ross on the South Polar expedition and
Huxley went on the Rattlesnake to the Australian Barrier Reef. The
Russians were also active, one of the more famous travellers being
Kotzebue, who was accompanied on one of his two voyages (1823-
6) round the world by Chamisso and Eschscholtz. In the early part of
this century the Americans were also enterprising, the work of Dana
being perhaps the most noteworthy. It would require several pages
to mention even the names of the naturalists who have had their
years of wandering, and have added their pages and sketches to the
book of the world’s fauna and flora, but such an enumeration would
serve no useful purpose here.
There is, however, one form of zoological exploration which
deserves a chapter to itself, that is the exploration of the Deep Sea.
Several generations of marine zoologists had been at work before a
zoology of the deep sea was dreamed of even as a possibility. It is
true that in 1818 Sir John Ross had found a star-fish (Astrophyton) at
a depth of 800-1000 fathoms, but this was forgotten; and in 1841
Edward Forbes dredged to no purpose in fairly deep water in the
Ægean Sea. Indeed those who thought about the great depths at all
deemed it unlikely that there could be life there, and if it had not
been for the practical affair of laying the ocean cables, we might
possibly have been still in ignorance of the abyssal fauna.
But the cables had to be laid—no easy task—and it became
important to know at least the topography of the depths. Cables
broke, too, and had to be fished up again, and when that which ran
between Sardinia and Algiers was lifted, in 1860, from a depth of 60-
1000 fathoms, no less than 15 different species of animals were
found on it. This was a discovery to fire enthusiasm, and Britain led
the way in following it up. In 1868 Wyville Thomson began his
explorations on the Lightning, and proved that most of the types of
backboneless animals were represented at depths of at least 600
fathoms. Soon followed the similar cruise of the Porcupine, famous
inter alia for the discovery of Bathybius, which many sceptics regard
as a mare’s nest. From various quarters the quest after the deep-sea
fauna began to be prosecuted.
It is now more than a score of years since the world-famous
Challenger sailed from Portsmouth with Wyville Thomson, Moseley,
John Murray, and Willemoes-Suhm as naturalists. During three and
a half years the explorers cruised over 68,900 nautical miles,
crossed the Atlantic no less than five times, reached with the long
arm of the dredge to depths equal to reversed Himalayas, raised
treasures of life from over 500 stations, and brought home spoils
over which the savants of Europe have hardly ceased to be busy,
and the records of which, now completed under Dr. Murray’s
editorship, form a library of about forty huge volumes.
The Challenger expedition was important not only in itself, but in the
wave of scientific enthusiasm which it raised. From Germany went
forth the Gazelle; Norway sent the Vöringen to Spitzbergen; America
has despatched the Tuscarora, the Blake, and the Albatross; from
Sweden the Vega and the Sophia sailed to Arctic seas: Count
Liechtenstein’s yacht Hertha explored Adria; the Prince of Monaco’s
Hirondelle darted hither and thither; the French sent forth the
Travailleur and Talisman; the Italians the Vettor Pisani and
Washington; Austria and Hungary organized the Poli for work in the
Mediterranean; the Germans again have recently specialized in
investigating the Plankton, or surface-life of the ocean; and so, with a
range even wider than we have indicated, the wave of enthusiasm
has spread, one of the latest barques which it has borne being the
Prince of Monaco’s, which was specially built for marine exploration.
Specialism in travelling has, of course, gone much further. Thus to
cite only three examples, we have Semper’s zoological work on the
Philippines, the researches of the Sarasins in Ceylon, and the first
results of Semon’s recent visit to Australasia, all of them passing far
beyond records of zoological exploration into monographs on the
structure and development of characteristic members of the fauna of
these countries. And it is no exaggeration to say that private
enterprise, Royal Society subsidies, British Association grants, and
the like have sent scores of naturalists from Britain half round the
world in order to solve special problems, as to the larva of a worm,
for instance, or as to the bird-fauna of some little island.
V. The Biological Type. In some ways the most important scientific
journey ever made was Darwin’s voyage on the Beagle. It was the
Columbus-voyage of zoology. There is a great deal to be said for the
Wanderjahre of the old students, for to have time to think is one of
the conditions of intellectual progress. Not that the Beagle voyage
was one of idleness, but it gave Darwin, at the age of twenty-two, a
wealth of impressions and some measure of enforced leisure
wherein to gloat intellectually over what he saw. He has said, indeed,
that various sets of facts observed on his voyage, such as the aspect
of the Galapagos Islands, started him on paths of pondering which
eventually led to his theory of the origin of species.
We take Darwin as the type of the biological, or, we may almost say,
evolutionist travellers; but he must share this position with his
magnanimous colleague, Alfred Russel Wallace, whose journeyings
were more prolonged and not less fruitful. Before Darwin the
naturalist-travellers had been, for the most part, describers,
systematists, and analysts, and it goes without saying that such work
is indispensable, and must continue; but in the light of the conception
of evolution all things had become new; the present world of life was
henceforth seen as a stage in a process, as a passing act in a
drama, not merely as a phantasmagoria to be admired and pictured,
but as a growth to be understood.
It is within this group of biological travellers, which includes such
men as Bates and Belt, that we must also place Brehm. For although
he perhaps had not the firmness of grasp or the fineness of touch
necessary for the successful handling of the more intricate biological
problems, especially those which centre around the factors of
evolution, he had unusual power as an observer of the habits of
animals. His contributions, which must be judged, of course, from his
great Tierleben,[A] as well as from his popular lectures, were rather
to the old natural history than to biology in the stricter sense. His
works show that he was as much interested in men as in beasts, that
he was specially an ornithologist, that he was beneath the naturalist
a sportsman; but so scores of other travellers have been. His
particular excellence is his power of observing and picturing animal
life as it is lived in nature, without taking account of which biology is
a mockery and any theory of evolution a one-sided dogma.
[A] This well-known treasure-house of Natural History appeared
originally in 1863-69 in six big volumes, which have since
increased to ten. Even the first edition took a foremost place
among similar works on the Natural History of Animals. With a
wealth of personal observation on the habits of animals in their
native haunts, it combined the further charm of very beautiful
pictorial illustration.
Let us now bring together briefly the outstanding facts of this
historical outline.
In early days men followed their wandering herds or pursued their
prey from region to region, or were driven by force of competition or
of hunger to new lands. Many of the most eventful journeys have
been among those which had to be taken.
I. Gradually, intellectual curiosity rather than practical need became
the prompter, and men travelled with all manner of mixed aims
seeking what was new. When they returned they told travellers’
tales, mostly in as good faith as their hunting ancestors had done in
the caves of a winter night, or as the modern traveller does after
dinner still. We pass insensibly from Herodotus to Marco Polo, from
“Sir John Maundeville” to Mr. X. Y. Z., whose book was published
last spring. This is the type romantic.
II. But when science shared in the renaissance there ensued the
extraordinary industry of the encyclopædist school, with which many
naturalist-travellers were associated. Some of these were great men
—perhaps Gesner was greatest of all—but all had the defects of
their qualities. They gathered into stackyards both wheat and tares,
and seldom found time to thrash. The type survives afield in the
mere collector, and its degenerate sedentary representatives are
called compilers.
III. Just as Buffon represents the climax of the encyclopædists, and
is yet something more, for he thrashed his wheat, so Humboldt,
while as ambitious as any encyclopædist traveller, transcended them
all by vitalizing the wealth of impressions which he gathered. He was
the general naturalist-traveller, who took all nature for his province,
and does not seem to have been embarrassed. Of successful
representatives of this type there are few, since Darwin perhaps
none.
IV. Meanwhile Linnæus had brought order, Cuvier had founded his
school of anatomists, Haller had re-organized physiology, the
microscope had deepened analysis, and zoology came of age as a
specialism. Henceforth travellers’ tales were at a discount; even a
Humboldt might be contradicted, and platitudinarian narratives of a
voyage round the world ceased to find the publisher sympathetic or
the public appetized. The naturalist-traveller was now a zoologist, or
a botanist, or an ornithologist, or an entomologist; at any rate, a
specialist. But it was sometimes found profitable to work in
companies, as in the case of the Challenger expedition.
V. Lastly, we find that on the travellers, too, “evolution” cast its spell,
and we have Darwin and Wallace as the types of the biological
travellers, whose results go directly towards the working out of a
cosmology. From Bates and Belt and Brehm there is a long list down
to Dr. Hickson, The Naturalist in Celebes, and Mr. Hudson, The
Naturalist in La Plata. Not, of course, that most are not specialists,
but the particular interest of their work is biological or bionomical.
I have added to this essay a list of some of the most important works
of the more recent naturalist-travellers with which I am directly
acquainted, being convinced that it is with these that the general,
and perhaps also the professional student of natural history should
begin, as it is with them that his studies must also end. For, not only
do they introduce us, in a manner usually full of interest, to the
nature of animal life, but they lead us to face one of the ultimate
problems of biology—the evolution of faunas.

II.
Alfred Edmund Brehm (1829-1884) was born at Unter-Renthendorf
in Sachsen-Weimar, where his father—an accomplished ornithologist
—was pastor. Brought up among birds, learning to watch from his
earliest boyhood, accompanying his father in rambles through the
Thuringian forest, questioning and being questioned about all the
sights and sounds of the woods, listening to the experts who came to
see the famous collection in the Pfarr-haus, and to argue over
questions of species with the kindly pastor, young Brehm was almost
bound to become a naturalist. And while the father stuffed his birds
in the evenings the mother read aloud from Goethe and Schiller, and
her poetic feeling was echoed in her son. Yet, so crooked are life’s
ways, the youth became an architect’s apprentice, and acted as
such for four years!
But an opportunity presented itself which called him, doubtless most
willing, from the desk and workshop. Baron John Wilhelm von Müller,
a keen sportsman and lover of birds, sought an assistant to
accompany him on an ornithological expedition to Africa, and with
him the youth, not yet out of his teens, set forth in 1847. It was a
great opportunity, but the price paid for it was heavy, for Brehm did
not see his home again for full five years, and was forced to bear
strains, to incur responsibilities, and to suffer privations, which left
their mark on him for life. Only those who know the story of his
African journeys, and what African travel may be with repeated
fevers and inconsiderately crippled resources, can adequately
appreciate the restraint which Brehm displays in those popular
lectures, here translated, where there is so much of everything but
himself.
After he returned, in 1852, rich in spoils and experience, if otherwise
poor, he spent several sessions at the universities of Jena and
Vienna. Though earnestly busy in equipping himself for further work,
he was not too old to enjoy the pleasures of a student life. When he
took his doctor’s degree he published an account of his travels
(Reiseskizzen aus Nordostafrica. Jena, 1855, 3 vols.).
After a zoological holiday in Spain with his like-minded brother
Reinhold—a physician in Madrid—he settled for a time in Leipzig,
writing for the famous “Gartenlaube”, co-operating with Rossmässler
in bringing out Die Tiere des Waldes, expressing his very self in his
Bird-Life (1861), and teaching in the schools. It was during this
period that he visited Lapland, of whose bird-bergs the first lecture
gives such a vivid description. In 1861 he married Matthilde Reiz,
who proved herself the best possible helpmeet.
In 1862, Brehm went as scientific guide on an excursion to Abyssinia
undertaken by the Duke of Coburg-Gotha, and subsequently
published a characteristic account of his observations Ergebnisse
einer Reise nach Habesch: Results of a Journey to Abyssinia
(Hamburg, 1863). On his return he began his world-famous
Tierleben (Animal Life), which has been a treasure-house to so
many naturalists. With the collaboration of Professors Taschenberg
and Oscar Schmidt, he completed the first edition of this great work,
in six volumes, in 1869.
Meanwhile he had gone to Hamburg as Director of the Zoological
Gardens there, but the organizing work seems to have suited him ill,
and he soon resigned. With a freer hand, he then undertook the
establishment of the famous Berlin Aquarium, in which he partly
realized his dream of a microcosmic living museum of nature. But,
apart from his actual work, the business-relations were ever irksome,
and in 1874 he was forced by ill-health and social friction to abandon
his position.
After recovery from serious illness he took up his rôle as popular
lecturer and writer, and as such he had many years of happy
success. A book on Cage Birds (1872-1876), and a second edition of
the Tierleben date from this period, which was also interrupted by his
Siberian journeys (1876) and by numerous ornithological
expeditions, for instance to Hungary and Spain, along with the
Crown Prince Rudolph of Austria. But hard work, family sorrows, and
finally, perhaps, the strain of a long lecturing tour in America aged
Brehm before his time, and he died in 1884.
For these notes I am indebted to a delightful appreciation of Brehm
which Ernest Krause has written in introduction to the third edition of
the Tierleben, edited by Pechuel-Loesche, and as regards the
naturalist’s character I can only refer to that essay. As to his
published work, however, every naturalist knows at least the
Tierleben, and on that a judgment may be safely based. It is a
monumental work on the habits of animals, founded in great part on
personal observation, which was always keen and yet sympathetic. It
is a classic on the natural history of animals, and readers of Darwin
will remember how the master honoured it.
Doubtless Brehm had the defects of his qualities. He was, it is said,
too generous to animals, and sometimes read the man into the beast
unwarrantably. But that is an anthropomorphism which easily besets
the sympathetic naturalist. He was sometimes extravagant and
occasionally credulous. He did not exactly grip some of the subjects
he tackled, such as, if I must specify, what he calls “the monkey-
question”.
It is frankly allowed that he was no modern biologist, erudite as
regards evolution-factors, nor did he profess to attempt what is
called zoological analysis, and what is often mere necrology, but his
merit is that he had seen more than most of us, and had seen, above
all, the naturalist’s supreme vision—the vibrating web of life. And he
would have us see it also.

III.
The success of the pictures which Brehm has given us—of bird-
bergs and tundra, of steppes and desert, of river fauna and tropical
forest—raises the wish that they had been complete enough to
embrace the whole world. As this ideal, so desirable both from an
educational and an artistic standpoint, has not been realized by any
one volume, we have ventured to insert here a list of some more or
less analogous English works by naturalist-travellers, sportsmen,
and others—
Adams, A. Leith. Notes of a Naturalist in the Nile Valley
and Malta (Edinburgh, 1870).
Agassiz, A. Three Cruises of the “Blake” (Boston and New
York, 1888).
Baker, S. W. Wild Beasts and their Ways: Reminiscences
of Europe, Asia, Africa, and America (London, 1890).
Bates, H. W. Naturalist on the Amazons (6th Ed. London,
1893).
Belt, T. Naturalist in Nicaragua (2nd Ed. London, 1888).
Bickmore, A. S. Travels in the East Indian Archipelago
(1868).
Blanford, W. T. Observations on the Geology and Zoology
of Abyssinia (London, 1870).
Bryden, H. A. Gun and Camera in Southern Africa
(London, 1893). Kloof and Karroo (1889).
Burnaby, F. A Ride to Khiva (8th Ed. London, 1877).
Buxton, E. N. Short Stalks, or Hunting Camps, North,
South, East, and West (London, 1893).
Chapman, A. and C. M. Buck. Wild Spain (London, 1892).
Cunningham, R. O. Notes on the Natural History of the
Straits of Magellan (Edinburgh, 1871).
Darwin, C. Voyage of the “Beagle” (1844, New Ed.
London, 1890).
Distant, W. L. A Naturalist in the Transvaal (London,
1892).
Drummond, H. Tropical Africa (London, 1888).
Du Chaillu, P. B. Explorations and Adventures in
Equatorial Africa (London, 1861). Ashango Land (1867).
Eha. A Naturalist on the Prowl, or in the Jungle (London,
1894).
Forbes, H. O. A Naturalist’s Wanderings in the Eastern
Archipelago (London, 1885).
Guillemard. Cruise of the “Marchesa” (London, 1886).
Heilprin, A. The Bermuda Islands (Philadelphia, 1889).
Hickson, S. J. A Naturalist in North Celebes (London,
1889).
Holub, Emil. Seven Years in South Africa (1881).
Hudson, W. H. The Naturalist in La Plata (London, 1892).
Idle Days in Patagonia (London, 1893).
Humboldt, A. von. Personal Narrative of Travels to the
Equinoctial Regions of America. Views of Nature (Trans.
1849). Cosmos (Trans. 1849-58).
Johnston, H. H. Kilima Ndjaro Expedition (1885).
Kingsley, C. At last! A Christmas in the West Indies
(1889).
Lumholtz. Among Cannibals (London, 1889).
Moseley, H. N. Notes by a Naturalist on the “Challenger”
(London, 1879. New Ed. 1892).
Nordenskiöld, A. E. Voyage of the “Vega” (London, 1881).
Oates, F., Ed. by C. G. Oates. Matabele Land, the Victoria
Falls, a Naturalist’s Wanderings in the Interior of South
Africa (1881).
Phillipps-Wolley. Big-Game Shooting (Badminton Libr.
London, 1893).
Rodway, J. In the Guiana Forest (London, 1894). British
Guiana (London, 1893).
Roosevelt, Th., and G. B. Grinell. American Big-Game
Hunting (Edinburgh, 1893).
Schweinfurth, G. The Heart of Africa (1878).
Seebohm, H. Siberia in Europe (London, 1880), Siberia in
Asia (London, 1882).
Selous, F. C. A Hunter’s Wanderings (1881). Travel and
Adventure in South-East Africa (London, 1893).
Sibree, Rev. J. The Great African Island (1879).
Solymos, B. (B. E. Falkenberg). Desert Life (London,
1880).
Stanley, H. M. How I Found Livingstone (1872, New Ed.
1885). The Congo (1885). Through the Dark Continent
(1890). In Darkest Africa (1890).
Swayne, H. G. C. Seventeen Trips through Somaliland
(London, 1895).
Tennent, J. E. Natural History of Ceylon (London, 1861).
Thomson, Wyville. The Depths of the Sea (London, 1873).
Narrative of the Voyage of the “Challenger” (1885). And, in
this connection, see S. J. Hickson. Fauna of the Deep Sea
(London, 1894).
Tristram, H. B. The Land of Israel (1876). The Land of
Moab (1873). The Great Sahara (1860).
Wallace, A. R. Malay Archipelago (London 1869). Tropical
Nature (1878). Island Life (1880). Travels on the Amazon
and Rio Negro (1889).
Waterton, Ch. Wanderings in South America (Ed. by J. G.
Wood, 1878).
Woodford, C. M. Naturalist among the Head-hunters
(London, 1890).
 FROM

NORTH POLE TO EQUATOR.

 THE BIRD-BERGS OF LAPLAND.
“When the Creator of the worlds had made the earth, best loved of
all, and was rejoicing in His perfect work, the devil was seized with a
desire to bring it all to nought. Not yet banished from heaven, he
lived among the archangels in the abodes of the blessed. Up to the
seventh heaven he flew, and, seizing a great stone, hurled it with
might down on the earth exulting in the beauty of its youth. But the
Creator saw the ruthless deed, and sent one of His archangels to
avert the evil. The angel flew even more swiftly than the stone to the
earth beneath, and succeeded in saving the land. The huge stone
plunged thundering into the sea, and hissing waves flooded all the
shores for many a mile. The fall shattered the crust of the stone, and
thousands of splinters sank on either side, some disappearing into
the depths, and some rising above the surface, bare and bleak like
the rock itself. Then God took pity, and in His infinite goodness
resolved to clothe even this naked rock with life. But the fruitful soil
was all but exhausted in His hand; there remained scarce enough to
lay a little here and there upon the stone.”
So runs an ancient legend still current among the Lapps. The stone
which the devil threw is Scandinavia; the splinters which fell into the
sea on either side are the skerries which form a richly varied wreath
around the peninsula. The rents and cracks in the rock are the fjords
and the valleys; the sprinkling of life-giving soil which fell from the
gracious Creator’s hand forms the few fertile tracts which
Scandinavia possesses. To appreciate the full depth and meaning of
the childish story one must one’s self have visited Scandinavia, and
especially Norway, have steered a boat among the skerries, and
have sailed round the country from the extreme south to the farthest
north. Marvellous, indeed, is the country; marvellous are its fjords;
still more marvellous is the encircling wreath of islands and reefs.
Scandinavia is an alpine country like Switzerland and the Tyrol, yet it
differs in a hundred ways from both of these. Like our Alps it has