Professional Documents
Culture Documents
Ebffiledoc 7220
Ebffiledoc 7220
Ebffiledoc 7220
https://textbookfull.com/product/handbook-of-adhesion-
technology-2nd-edition-lucas-f-m-da-silva/
https://textbookfull.com/product/individual-behaviors-and-
technologies-for-financial-innovations-wesley-mendes-da-silva/
https://textbookfull.com/product/postcolonial-maghreb-and-the-
limits-of-ir-jessica-da-silva-c-de-oliveira/
https://textbookfull.com/product/networks-in-systems-biology-
applications-for-disease-modeling-fabricio-alves-barbosa-da-
silva/
Zika Virus and Diseases From Molecular Biology to
Epidemiology 1st Edition Suzane R. Da Silva
https://textbookfull.com/product/zika-virus-and-diseases-from-
molecular-biology-to-epidemiology-1st-edition-suzane-r-da-silva/
https://textbookfull.com/product/fundamentals-of-renewable-
energy-processes-4th-edition-aldo-vieira-da-rosa/
https://textbookfull.com/product/industrial-applications-of-
adhesives-1st-international-conference-on-industrial-
applications-of-adhesives-lucas-f-m-da-silva/
https://textbookfull.com/product/quantum-computing-by-
practice-2nd-edition-vladimir-silva/
https://textbookfull.com/product/all-grown-up-1st-edition-
larissa-de-silva-de-silva/
INTERNATIONAL ASSOCIATION FOR HYDRO-ENVIRONMENT ENGINEERING AND RESEARCH
IAHR MONOGRAPH
Fluvial Processes
2nd Edition
2nd Edition
Peter A. Davies
University of Dundee, UK
(Series Editor)
Table of contents
1 Fundamentals 1
1.1 Sediment transport 1
1.2 Turbulent flow 2
1.2.1 General 2
1.2.2 Vertical distributions of shear stress and flow velocity 3
1.2.3 Average flow velocity 7
1.2.4 Resistance equation 7
1.3 Two-dimensional two-phase motion 8
1.4 Bed-load rate; Bagnold’s formula 11
1.5 Vector forms of the bed-load rate 13
1.6 Suspended-load rate 15
1.7 Sediment transport continuity equation 17
1.8 Additional remarks 21
Problems 22
References 25
2 Bed forms 29
2.1 General 29
2.2 Origin of bed forms 29
2.2.1 Dunes and bars 30
2.2.2 Ripples 45
2.2.3 Clarification of “uniform flows’’ 48
2.2.4 Special aspects 48
2.2.5 Additional remarks: the ongoing debate
on the origin of bed forms 49
2.3 Geometry and existence regions of dunes and ripples 50
2.3.1 Introductory considerations 50
2.3.2 Geometric characteristics of dunes 51
2.3.3 Geometric characteristics of ripples 57
2.3.4 Existence regions of dunes and ripples 60
2.4 Geometry and existence regions of bars 61
viii Table of contents
Much research has been carried out on fluvial processes and river morphodynamics
since the publication of the 1st edition of this book. In the preparation of this version, I
tried as much as possible to incorporate the results of such research, while adhering to
the central theme of the book as stated in the ‘Preface to the 1st edition’, and without
losing its coherence and logical sequence.
An effort was made to incorporate also comments on the book kindly provided
throughout the years by several colleagues and students.
All chapters were updated and revised. The book consists of seven chapters. Just
like in the 1st edition, the first chapter is devoted to the basics of turbulent flow and
sediment transport. Chapter 2 deals with the origin, existence regions and geometric
properties of bed forms (ripples, dunes, alternate and multiple bars). Chapter 3 con-
cerns the flow over undulated beds. Chapters 2 and 3 are a significantly expanded
version of the original Chapter 2. The extension enabled the inclusion of: 1- a more
comprehensive presentation and discussion of works focusing on the characteristic
scales and dynamics of large-scale turbulence coherent structures; 2- a new section
on the geometric properties of alternate bars; and 3- a new section on the internal
structure of flow over dunes. The remaining chapters follow the same order as in the
1st edition. Accordingly, Chapter 4 concerns the regime concept, and focuses on the
reason for the formation of regime channels and also the computation of their per-
tinent geometric characteristics. In Chapter 5 we provide an analysis on how regime
processes take place, with the time-development of meandering and braiding streams
being discussed in the light of the regime-trend. The formation of deltas is also briefly
considered. Chapter 6 deals with the geometry and mechanics of meandering streams.
This chapter too is considerably expanded in comparison to its counterpart in the 1st
edition, incorporating a more detailed description of the mechanics of meandering
flows and their interaction with the deformable bed, as derived from recent labora-
tory and numerical experiments. Finally, Chapter 7 deals with the computation of bed
deformation and planimetric evolution of meandering streams. All chapters, with the
exception of Chapter 2, include a related set of problems. However, the problems at
the end of Chapter 3, which concern primarily the determination of the resistance
factor and sediment transport rate of flow past undulated beds, rely on methods and
equations regarding the geometric properties of bed forms introduced in Chapter 2 –
and as such provide ample opportunity to practice the application to real problems
of the methods introduced in Chapter 2. The supplementary material, consisting of
the solutions manual for the sets of problems in the book and the computer programs
xii Preface to the 2nd edition
A stream flowing in alluvium deforms its bed surface so as to form ripples, dunes,
bars, etc., and, in many instances, it deforms its channel as a whole so as to create, in
plan view, meandering or braiding patterns. One can say that, in general, an alluvial
stream and its deformable boundary undergo a variety of fluvial processes which lead
to the emergence of a variety of alluvial forms.
This monograph concerns the understanding and quantitative formulation of fluvial
processes and the associated alluvial forms. It is designed for researchers and graduate
students of hydraulic engineering, water resources, and the related branches of earth
sciences. However, it may have some appeal to practicing professionals as well.
The central theme of the book is that the initiation of periodic (in flow direc-
tion) large-scale alluvial forms is due to large-scale turbulence, the subsequent time-
development of the so-initiated alluvial forms being guided by the regime trend.
The text is of a deductive nature: the content of any chapter presupposes the knowl-
edge of the contents of preceding chapters; hence, the text might not appear compre-
hensible if it is not read in sequence. Dimensional methods are extensively used, and
the importance of the “agreement with experiment’’ is stressed. The stability-approach
is not a part of the research field of the authors, and it is not used in this text. (The
reader interested in the study of fluvial processes by stability is referred to the excellent
works of G. Seminara, G. Parker, M. Tubino, T. Hayashi).
It was gratifying for the senior author to learn that some of his previous books
are used in graduate courses in some universities. Considering this, the authors have
tried to make this text “student-oriented’’. Thus each chapter is supplemented by a
set of related problems, and, wherever appropriate, some topics for future research
are suggested. The FORTRAN programs for the computation of resistance factor and
regime channels are also included.
The representation of an experimental point-pattern by an appropriate equation
which can be used for computational purposes (in short, by a “comp-eq.’’) is always
desirable. Hence a number of “comp-eqs.’’ is suggested in this text. These equations
have no claim other than that their graphs pass through the midst of the respective
point-patterns.
The first chapter of this book presents the basics of turbulent flow and sediment
transport; the second deals with the bed forms and flow resistance. These two (classical)
chapters are but the updated and revised extensions of their counterparts in the book
“River Mechanics’’ (M.S. Yalin, Pergamon Press, 1992). Chapter 3 concerns the regime
concept and its thermodynamic formulation. The development of meandering and
xiv Preface to the 1st edition
braiding streams in the light of the regime trend is discussed in Chapter 4, where
the formation of deltas is also considered. Much of the ongoing research in fluvial
hydraulics is related to meandering streams, and Chapter 5 is devoted exclusively to
the study of the geometry and mechanics of these streams. The computation of flow,
bed deformation, and migration-expansion of meandering streams forms the topic of
Chapter 6.
The authors are grateful to Prof. H. Scheuerlein (Chairman Division III – IAHR)
for his continual and enthusiastic help and encouragement. They feel also indebted to
Prof. G. Di Silvio and Prof. M. Jaeggi, for reviewing this manuscript and making a
number of valuable suggestions. For any imperfection that may still remain, it is solely
the authors who are to be blamed.
The authors would like to express their thanks to Dr. T. J. Harris, Dean of the Faculty
of Applied Science, Queen’s University, and to Dr. D. Turcke, Head of the Department
of Civil Engineering, Queen’s University, for their generous financial support. Their
thanks go also to Larry Harris, Graphic Design Unit, Queen’s University.
In this text the term alluvium is used as an abbreviation for cohesionless granular
material or medium, alluvial stream being the flow whose channel, or only the channel
bed, is alluvial.
The natural alluvial streams are usually “wide’’: their aspect ratio (width-to-depth
ratio) usually is larger than ≈10, say. The aspect ratio increases with the flow rate,
and in large natural rivers this ratio is often a three-digit number. The (turbulent) flow
in such rivers is almost always sub-critical.
In physical sciences, a natural process is studied with the aid of its idealized coun-
terpart, where the “natural arbitrariness’’ is removed, and the process is brought into
a form suitable for mathematical treatment; a fluvial process is no exception.
The present text “aims’’ at large alluvial streams (rivers). Hence, it is assumed
throughout this text that the idealized alluvial stream (ideal river) is wide, and that its
flow is turbulent and sub-critical.
The studies of regime channels, and of meandering and braiding streams are carried
out nowadays for a constant (representative) flow rate Q – and so it is done in the
following. (The authors’ views on the representative Q are summarized in Section 4.7).
List of symbols
1. General
fA dimensional function determining a quantity A
Φ A , φ A , ΨA , ψA dimensionless functions determining a quantity A
α, α , β, β dimensionless coefficients (not necessarily constants) in the
expression of a quantity
≈ approximately equal to, comparable with
∼ proportional to (proportionality factor may not be a constant)
∇ nabla-operator (“del’’)
2. Average values
A vertically-averaged value of a quantity A
Am cross-sectional average value of a quantity A
Aav channel average value of a quantity A
3. Subscripts
a, O mark the values of a quantity at the apex- and crossover-sections of
a meandering stream, respectively
b marks the value of a quantity at the bed, or related to the bed
cr marks the value of a quantity corresponding to the initiation of
sediment transport (to the “critical stage’’)
max marks the maximum value of a quantity
min marks the minimum value of a quantity
R marks the regime value of a quantity
0 marks, as a rule, the value of a quantity at t = 0; exceptions are τ0 ,
cM0 , and θ0
4. Coordinates
t time
x direction of rectilinear flow; also general direction of meandering flow
y direction horizontally perpendicular to x
z vertical direction in general (also elevation of a point – see
“5. Pertinent quantities’’)
z+ dimensionless “wall’’ coordinate (=υ∗ z/ν)
xviii List of symbols
5. Pertinent quantities
A mechanical quantity in general; also area
a 1 , a2 consecutive apex-sections (in Chapter 4)
ai , ai+1 consecutive apex-sections (in Chapters 6 and 7)
A∗ energy-related property of flow (subjected to minimization during
the regime channel formation); also coefficient in the bed-load
formula of H.A. Einstein (Problem 1.11)
B flow width at the free surface
Bc width of the central region of the stream cross-section
Bs roughness function
c total dimensionless (Chézy) resistance factor
cf pure friction component of c
c∆ bed form component of c
cM total local dimensionless resistance factor of a meandering flow
cM0 local bed resistance component of cM (bed form effect included)
C local dimensionless volumetric concentration of suspended-load
C the value of C at z =
CV control volume
D typical grain size (usually D50 )
ei internal energy per unit fluid mass at a space point m
ek kinetic energy per unit fluid mass at a space point m
ep pressure work per unit fluid mass at a space point m
e total energy per unit fluid mass at a space point m (e = ei + ek + ep )
Ei internal energy of the fluid in Sys or CV
E total energy of the fluid in Sys or CV
e V , eH eddy-burst forming eddies of vertical and horizontal turbulence,
respectively
Fi energy transfer per unit time through a cross-section Ai (energy flux
plus the displacement work)
g acceleration due to gravity
h flow depth
iα unit vector in the direction of α
J longitudinal free surface slope
J radial free surface slope
List of symbols xix
ρ fluid density
ρs grain density
σ sinuosity of a meandering flow (σ = L/ΛM ); also area of the side
surface of an imaginary vertical prism of height h and base area
A (Chapter 1)
τ shear stress
τ0 magnitude of the bed shear stress vector τ0
χ magnitude of the local “side stress’’ resultant vector χ
φr angle of repose
ω deviation angle (angle between the streamlines s and coordinate lines
l of a meandering flow)
ωc value of ω along the centreline of a meandering flow
ωz vertical component of vorticity (Chapter 2)
6. Dimensionless combinations
Fr flow Froude number (Fr = V 2 /(gh), where V is a typical flow velocity
(V = u, um , uav , ..., etc.))
Re flow Reynolds number (Re = Vh/ν, where V is as above)
Re∗ roughness Reynolds number (Re∗ = υ∗ ks /ν)
X grain size Reynolds number (X = υ∗ D/ν)
Y mobility number (Y = ρυ∗2 /(γs D))
Z relative depth (Z = h/D)
W density ratio (W = ρs /ρ)
Ξ3 material number (Ξ 3 = X 2 /Y = γs D3 /(ρν 2 ))
η∗ relative flow intensity (η∗ = Y/Ycr )
1/2
φ Einstein’s dimensionless transport rate (φ = ρ 1/2 qs /(γs D3/2 ))
N dimensionless specific flow rate (N = Q/(BDυ∗cr ))
7. Abbreviations
[CD] L/2-long convergence-divergence flow region
CS turbulence coherent structure; also control surface (Chapter 4)
LSHCS large-scale horizontal coherent structure (occasionally referred
to as HCS)
LSVCS large-scale vertical coherent structure (occasionally referred
to as VCS)
Chapter 1
Fundamentals
Any flow-induced deformation of an alluvial channel (or only of its bed) is by means
of the grain motion en mass, i.e. by means of sediment transport. Hence we start this
text by considering some aspects of sediment transport and related topics.
The grains forming the boundary of an alluvial stream have a finite weight (in fluid)
and a finite coefficient of friction. Consequently they cannot be brought into motion
by the flow if the shear stress τ0 acting at a point on the flow boundary (Figure 1.1)
is less than a certain “critical’’ value (τ0 )cr (corresponding to that point). This means
that the sediment can be transported only over that part A B of the flow boundary
AA B B where
τ0
η∗ = > 1. (1.1)
(τ0 )cr
In this text the ratio η∗ will be referred to as relative tractive force or relative flow
intensity. Only the grains forming the uppermost grain layer of the flow boundary can
be detached and transported by flow (in the flow direction x). The detachment is due
to τ0 ; the transport, to the longitudinal flow velocities u.
If η∗ is smaller than a certain value, η∗1 say (1 < η∗ < η∗1 ), then the grains are
transported by the deterministic “jumps’’ Pb in the neighbourhood of the bed (ks < z <
in Figure 1.2). This mode of grain transport is referred to as bed-load.
(1.1) boundary.
boundary.
(1.1) (1.1)
(1.1) boundary.
boundary. (1.1)
(1.1) (1.1)
(1.1)
Figure 1.1 Schematic representation of cross-sectional distribution of shear stress acting on the flow
boundary.
2 Fluvial processes
(1.1)
(1.1) (1.1)
(1.1)
(1.1)
(1.1)
(1.1) (1.1) (1.1) (1.1)
(1.1)
(1.1)
If η∗ is “large’’ (η∗ > η∗1 ), then some of the transported grains “diffuse’’, by turbu-
lence, into the remaining flow region < z < h, while some others are still transported
in the form of bed-load, i.e. within ks < z < . The transport of grains within < z < h,
along the probabilistic paths Ps , is referred to as suspended-load. It should thus be
clear that suspended-load is not a “substitute for’’, but an “addition to’’ the bed-load.
It appears that η∗1 must be a certain function of the grain size Reynolds number
υ∗ D/ν – which will be denoted in this text by X (see Section 1.3 later on). It has not
yet been settled what form this function must possess; for contemporary views on the
topic see, e.g., Cheng and Chiew (1999).
The total volume of grains passing through a flow cross-section per unit flow width
and per unit time is referred to as the specific
T and volumetric total transport rate qs .
This is the sum of the bed-load rate qsb and the suspended-load rate qss (each of which
is, of course, also specific and volumetric):
qs = qsb + qss (with qs ≥ qsb ; qs > qss ). (1.2)
The dimension of any qsi is
[qsi ] = [length] · [velocity]. (1.3)
1.2.1 General
The basic laws of sediment transport correspond to the simplest case of a steady and
uniform two-dimensional turbulent flow in a straight open-channel. In the present
context, “two-dimensional’’ means that the flow characteristics do not vary along
the third dimension y (perpendicular to the vertical (x; z)-planes). The simplest case
mentioned may sound as rather abstract and detached from reality. This is not so; the
cross-sections of wide natural alluvial streams (rivers) are, in their straight reaches,
nearly trapezoidal (see e.g. the (highly exaggerated along z) river cross-sections in
Figures 1.3a,b), while their variation in the flow direction x often is insignificant. In
such reaches, the flow in the (practically unaffected by the bank friction) “central
region’’ Bc (Figure 1.3a) approximates closely to a uniform two-dimensional flow.
Fundamentals 3
a) B=1480 m b) 0 20 40
EL.(m)
Be Horizontal
10 Scale (m)
z
5
hm =11 m
0
at Sabananueva
y Volga R. EL.(m) Sinu R. (Colombia)
10
B=840m
hm=4m 5
Figure 1.3 Examples of river cross-sections in straight reaches: (a)Volga and Syr Darya Rivers (adapted
from Velikanov 1995); (b) Sinu River (adapted from Monsalve and Silva 1983).
a) z, b) z
_umax
h h
-U
T
0.368h fi
ks
o T T
0
U
Figure 1.4
Figure 1.4 Vertical distribution of: (a) shear stress; (b) flow velocity.
In the following, up to the end of Section 1.5, it will be assumed that the mobile
bed surface is flat, and that it possesses a granular roughness ks which can be evaluated
(following Kamphuis 1974, Yalin 1977, 1992) as
ks ≈ 2D. (1.4)
32
11
10
91!,5
Bs 32 5,
8
7
Hyd. Transitionally . Fully .
6
Smooth
32
Rough .....
o Rough
5.5
o 32
502 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 32
log Re.
Figure 1.5 Plot of roughness function versus roughness Reynolds number (adapted from Schlichting
1968): experimental data (open and filled circles) and graph of Eq. (1.10) (solid red line).
For the derivation of Eqs. (1.5) and (1.6), see e.g. Schlichting (1968), Monin and
Yaglom (1971) and Yalin (1977). In these expressions
τ0 = γ Sh, (1.7)
τ0
υ∗ = = gSh (shear velocity), (1.8)
ρ
and1
As = eκBs . (1.9)
Here κ is the von Kármán constant. If the fluid is “clear’’, i.e. no sediment in suspen-
sion, then κ ≈ 0.4. The “roughness function’’ Bs = φB (Re∗ ), where Re∗ = υ∗ ks /ν is the
roughness Reynolds number, is determined by the experimental curve in Figure 1.5.
This curve is well represented, throughout its existence-region (0.2 < log Re∗ < 3.2),
by the comp-eq.
2.55
2.55
Bs = (2.5 ln Re∗ + 5.5)e−0.0705(ln Re∗ ) + 8.5 1 − e−0.0594(ln Re∗ ) (1.10)
introduced in da Silva and Bolisetti (2000). The solid line in Figure 1.5 is the graph of
Eq. (1.10).
Note that:
1 If Re∗ <≈ 5, then the turbulent flow is said to be in the hydraulically smooth
regime; if ≈5 < Re∗ <≈ 70, then it is in the transitionally rough regime; and if
1
If the two-dimensional flow is not uniform, then its free surface slope J would differ from the
bed slope S, and the value of τ0 would be given by τ0 = γ Jh ( = γ Sh).
Fundamentals 5
uu z+ 1Viscous sublayer
a; 1:a;a; Buller layer
1: 1:
.!: .!:
.!: -=0.15-0.2h
30.0
1: U
1: .9
0 CI 11.6
.c, !! -5.0
!! u
~
a; 6,
1:
.!: .Q
uu
Figure 1.6 Schematic representation of the velocity profile in a hydraulically smooth flow.
Re∗ >≈ 70, then it is in the fully rough regime (or it is a fully rough turbulent
flow).
2 For practical purposes, it is usually assumed that Eq. (1.6) represents the velocity
profile throughout the region z = zmin to h, where zmin is either equal to 11.6ν/υ∗
or ks , whichever is the larger. This approach is adopted throughout this book.
With regard to point 2 above, it should, however, be noted that in reality the
vertical velocity profile is more complex than implied by Eq. (1.6). Indeed, even though
generally regarded as a good approximation to the entire velocity profile (above zmin ),
in fact Eq. (1.6) is strictly valid within the inner flow region (z/h <≈ 0.15 to 0.2).
The deviations of the velocity profile from the logarithmic law in the outer region
(z/h >≈ 0.15 to 0.2) can be adequately taken into account by adding to Eq. (1.6) an
additional term, known as “wake function’’ φw (z/h). The most commonly invoked
wake function is that by Coles (1956), which can be expressed as φw (z/h) = (2Π/κ)·
sin2 (Π z/2h) (see also Hinze 1975). Here Π is the Coles’s wake strength parameter.
However, other formulae for the wake function, or modified versions of Coles’ original
function, are available in the literature (see e.g. Granville 1976, Krogstad et al. 1992,
and Guo et al. 2005).2
The conditions near the bed in the case of hydraulically smooth flows and the effect
of relative submergence in the case of rough beds deserve also some further discussion.
Under hydraulically smooth conditions, the flow in a thin region near the bed
consists of two layers, namely the viscous sublayer (z+ <≈ 5) and the buffer layer
(≈5 < z+ <≈ 30) (see Figure 1.6). Here z+ = υ∗ z/ν. Within the viscous sublayer, the
u-distribution is given by the linear form u/υ∗ = υ∗ z/ν. In the buffer layer, neither this
linear form nor the logarithmic law (Eq. (1.6)) hold. While direct expressions for the
u-distribution within the buffer layer are not available (Nezu and Nakagawa 1993,
2
Since the sediment is transported mainly in the lower fluid layers, the function φw (z/h) will be
disregarded in this text.
6 Fluvial processes
Outer layer
Logarithmic layer
Form-induced sublayer
Interfacial sublayer
Subsurface layer
Figure 1.7 Flow layers in a fully rough flow (after Nikora et al. 2001).
Nezu 2005), the just mentioned linear form provides a reasonable approximation
to the velocity profile up to z+ ≈ 11.6; while, on the other hand, the logarithmic law
reasonably approximates the velocity profile within ≈11.6 < z+ <≈ 30 (see e.g. Monin
and Yaglom 1971). This explains the reason for the statement “... where zmin is equal
to 11.6ν/υ∗ ...’’ in the point 2 above.
For the case of fully rough flows, it has since long been established that, as long as
h ks , the vertical velocity profile exhibits properties similar to those of hydraulically
smooth flows – in the sense that the profile follows the logarithmic law within the inner
region down to some distance from the bed (see e.g. Raupach et al. 1991, Nezu and
Nakagawa 1993, Jiménez 2004). Open-channel flows with small relative submergence
(small h/ks ), such as gravel rivers where the roughness elements can protrude the flow
up to the free surface, have also been the focus of considerable research especially
in more recent years (see e.g. Dittrich and Koll 1997, Nikora et al. 2001, Koll 2006,
McSherry et al. 2016, among others). It follows that fully rough flows can most appro-
priately be sub-divided into five layers (see Figure 1.7): outer layer, logarithmic layer,
form-induced sublayer (just above the roughness crests where the flow is influenced by
individual roughness elements and the velocity u acquires a linear profile), interfacial
sublayer (between roughness crests and throughs), and in the case of a permeable bed,
also a sub-surface layer (Nikora et al. 2001, Koll 2006). Accordingly, the flow can fur-
ther be sub-divided into three types: 1- flow with high relative submergence (h ks ),
exhibiting all the above layers (and in which the outer layer is similar to the outer layer
of hydraulically smooth flows, in the sense of Nezu and Nakagawa 1993); 2- flow
with small relative submergence (ks ≤ h < 2 to 5ks ), with the form-induced sublayer
as the upper flow region (and where therefore the logarithmic layer is suppressed);
and 3- flow with a partially inundated bed (h < ks ), with the interfacial sublayer as
the upper flow region. Clearly, Eq. (1.6) applies only to the first of these three types –
in which case for practical purposes, and as follows from point 2 above, Eq. (1.6) is
assumed to provide a reasonable approximation of the velocity profile from zmin = ks
to h. Figure 1.4b illustrates this case.
Another random document with
no related content on Scribd:
DAMSON JELLY.
Pour the juice from some damsons which have stood for a night in
a very cool oven, or been stewed in a jar placed in a pan of water;
weigh and put it into a preserving-pan with a pound and four ounces
of pearmains (or of any other fine boiling apples), pared, cored, and
quartered, to each pound of the juice; boil these together, keeping
them well stirred, from twenty-five to thirty minutes, then add the
sugar, and when it is nearly dissolved, continue the boiling for ten
minutes. This, if done with exactness, will give a perfectly smooth
and firm preserve, which may be moulded in small shapes, and
turned out for table. The juice of any good red plum may be used for
it instead of that of damsons.
To each pound clear damson-juice, 1-1/4 lb. pearmains (or other
good apples), pared and cored: 25 to 30 minutes. Sugar, 14 oz.: 10
minutes.
EXCELLENT DAMSON CHEESE.
Strip from their stalks some fine ripe black-cluster grapes, and stir
them with a wooden spoon over a gentle fire until all have burst, and
the juice flows freely from them; strain it off without pressure, and
pass it through a jelly-bag, or through a twice-folded muslin; weigh
and then boil it rapidly for twenty minutes; draw it from the fire, stir in
it until dissolved, fourteen ounces of good sugar, roughly powdered,
to each pound of juice, and boil the jelly quickly for fifteen minutes
longer, keeping it constantly stirred, and perfectly well skimmed. It
will be very clear, and of a beautiful pale rose-colour.
Juice of black-cluster grapes: 20 minutes. To each pound of juice,
14 oz. good sugar: 15 minutes.
Obs.—We have proved this jelly only with the kind of grape which
we have named, but there is little doubt that fine purple grapes of
any sort would answer for it well.
ENGLISH GUAVA.
Weigh six pounds of the fruit when it is quite ripe, but before the
frost has touched it; give each plum a cut as it is thrown into the
preserving-pan, and when all are done boil them from thirty-five to
forty minutes, taking out the stones as they rise to the surface, when
they are quite detached from the flesh of the fruit. Draw back the pan
from the fire, stir in two pounds of good sugar beaten to powder, and
boil the preserve quickly for fifteen minutes. The imperatrice plum is
of itself so sweet that this proportion of sugar makes with it a very
rich preserve.
Imperatrice plums (without the stalks) 6 lbs.: boiled 35 to 40
minutes. Sugar 2 lbs. (added after the stones are out): 15 minutes.
Obs.—Some slight trouble would be avoided by pressing the fruit
through a sieve after the first boiling; but we do not think the
marmalade would be improved by being freed from the skins of the
plums.
TO DRY IMPERATRICE PLUMS.
Gather the fruit in the middle of the day in very dry weather; strip
off the stalks, and have in readiness some perfectly clean and dry
wide-necked bottles; turn each of these the instant before it is filled,
with the neck downwards, and hold in it two or three lighted matches:
drop in the fruit before the vapour escapes, shake it gently down,
press in some new corks, dip the necks of the bottles into melted
resin, set them at night into an oven from which the bread has been
drawn six or seven hours at least, and let them remain until the
morning: if the heat be too great the bottles will burst. Currants,
cherries, damsons, greengages, and various other kinds of plums
will remain good for quite twelve months when bottled thus, if stored
in a dry place.
To steam the fruit, put the bottles into a copper or other vessel up
to their necks in cold water, with a little hay between and under them;
light the fire, let the water heat slowly, and keep it at the point of
gentle simmering until the fruit is sufficiently scalded. Some kinds will
of course require a much longer time than others. From half to three
quarters of an hour will be sufficient for gooseberries, currants, and
raspberries; but the appearance of all will best denote their being
done. When they have sunk almost half the depth of the bottles, and
the skins are shrivelled, extinguish the fire, but leave them in the
water until it is quite cold; then wipe and store the bottles in a dry
place. A bit of moistened bladder tied over corks is better than the
resin when the fruit is steamed.
APPLE JELLY.
Pare, quarter, core, and weigh some ripe but quite sound quinces,
as quickly as possible, and throw them as they are done into part of
the water in which they are to be boiled, as directed at page 456;
allow one pint of this to each pound of the fruit, and simmer it gently
until it is a little broken, but not so long as to redden the juice, which
ought to be very pale. Turn the whole into a jelly-bag, or strain the
liquid through a fine cloth, and let it drain very closely from it but
without the slightest pressure. Weigh the juice, put it into a delicately
clean preserving-pan, and boil it quickly for twenty minutes; take it
from the fire and stir in it, until it is entirely dissolved, twelve ounces
of sugar for each pound of juice, or fourteen ounces if the fruit should
be very acid, which it will be in the earlier part of the season; keep it
constantly stirred and thoroughly cleared from scum, from ten to
twenty minutes longer, or until it jellies strongly in falling from the
skimmer; then pour it directly into glasses or moulds. If properly
made, it will be sufficiently firm to turn out of the latter, and it will be
beautifully transparent, and rich in flavour. It may be made with an
equal weight of juice and sugar mixed together in the first instance,
and boiled from twenty to thirty minutes. It is difficult to state the time
precisely, because from different causes it will vary much. It should
be reduced rapidly to the proper point, as long boiling injures the
colour: this is always more perfectly preserved by boiling the juice
without the sugar first.
To each pound pared and cored quinces, 1 pint water: 3/4 to 1-1/2
hour. Juice, boiled 20 minutes. To each pound, 12 oz. sugar: 10 to
20 minutes. Or, juice and sugar equal weight: 20 to 30 minutes.
QUINCE MARMALADE.
Take the finest barberries without stones that can be procured, tie
them together in bunches of four or five sprigs, and for each half
pound of the fruit (which is extremely light), boil one pound of very
good sugar in a pint of water for twenty minutes, and clear it well
from scum; throw in the fruit, let it heat gently, and then boil from five
to seven minutes, when it will be perfectly transparent. So long as
any snapping noise is heard the fruit is not all done; it should be
pressed equally down into the syrup until the whole of the berries
have burst; and should then be turned into jars, which must be
covered with skin or two or three folds of thick paper, as soon as the
preserve is perfectly cold. The barberries thus prepared make a
beautiful garnish for sweet dishes, or for puddings.
Barberries, tied in bunches, 1-1/2 lb.; sugar 3 lbs.; water 1-1/2 pint:
20 minutes. Barberries boiled in syrup: 5 to 7 minutes.
BARBERRY JAM.
Strip the fruit from the stems, wash it in spring-water, drain, bruise
it slightly, and put it into a clean stone jar, with no more liquid than
the drops which hang about it. Place the jar in a pan of water, and
steam the fruit until it is quite tender: this will be in from thirty minutes
to an hour. Pour off the clear juice, strain, weigh, and boil it quickly
from five to seven minutes, with eighteen ounces of sugar to every
pound. For the marmalade, rub the barberries through a sieve with a
wooden spoon, and boil them quickly for the same time, and with the
same proportion of sugar as the jelly.
Barberries boiled in water-bath until tender; to each pound of juice,
1 lb. 2 oz. sugar: 5 minutes. Pulp of fruit to each pound, 18 oz.
sugar: 5 minutes.
Obs.—We have always had these preserves made with very ripe
fruit, and have found them extremely good; but more sugar may be
needed to sweeten them sufficiently when the barberries have hung
less time upon the trees.
ORANGE MARMALADE.
(A Portuguese Receipt.)
Rasp very slightly on a fine and delicately clean grater the rinds of
some sound Seville oranges; cut them into quarters, and separate
the flesh from the rinds; then with the small end of a tea, or egg-
spoon, clear it entirely from the pips, and from the loose inner skin
and film. Put the rinds into a large quantity of cold water, and change
it when they have boiled about twenty minutes. As soon as they are
perfectly tender lift them out, and drain them on a sieve; slice them
thin, and add eight ounces of them to each pound of the pulp and
juice, with a pound and a half of highly-refined sugar in fine powder;
boil the marmalade quickly for half an hour, skim it well, and turn it
into the jars. The preserve thus made will not have a very powerful
flavour of the orange rind. When more of this is liked, either leave a
portion of the fruit unrasped, or mix with the preserve some of the
zest which has been grated off, allowing for it its weight of sugar. Or
proceed thus: allow to a dozen Seville oranges two fine juicy lemons,
and the weight of the whole in sifted sugar, of excellent quality. With
a sharp knife cut through the rinds just deep enough to allow them to
be stripped off in quarters with the end of a spoon, and throw them
for a night into plenty of cold spring-water; on the following morning
boil them sufficiently tender to allow the head of a pin to pierce them
easily; then drain them well, let them cool, and scrape out the white
part of the rind, and cut the remainder into thin chips. In the mean
time have the pulp of the fruit quite cleared from the pips and film;
put it with the chips into a preserving-pan, heat them slowly, boil
them for ten minutes, draw the pan from the fire, and stir gradually
in, and dissolve the remainder of the sugar, and boil the preserve
more quickly for twenty minutes, or until it thickens and appears
ready to jelly. This mode, though it gives a little additional trouble, will
prevent the orange-chips from becoming hard, which they will
sometimes be if much sugar be added to them at first. The sugar first
broken into large lumps, is sometimes made into a very thick syrup,
with so much water only as will just dissolve it; the pulp and juice are