JOURNAL OF MORPHOLOGY 189313-325(1986)

Soft-Tissue Bone Interface: How Do Attachments of Muscles,

Tendons, and Ligaments Change During Growth?
A Light Microscopic Study
JACK R. HUROV
DeDartment of Orthodontics. School of Dental Medicine, Washington
Urhersity, S i Louis, Missouri 63110'

ABSTRACT This study addressed the problem of how soft structures main-
tain approximately the same relative positional relationships during long bone
growth. Attachments of the popliteus muscle, semitendinosus tendon, medial
collateral knee ligament, and extensor retinaculum were examined histologi-
cally in rabbits, aged 2-60 days, to determine the manner in which soft
structures attached to long bones during growth. Soft structures inserted
principally into fibrous periosteum or perichondrium in the age range studied.
However, an extensive collagen fiber framework within the cellular perios-
teum and perichondrium, present by at least 2 days of age, linked the fibrous
periosteum or perichondrium to subjacent bone or cartilage. Maturation of soft
tissue-bone interfaces was viewed from two related perspectives. The first
stressed temporal patterning of cartilage and bone differentiation. The second
emphasized incorporation of attachments of soft structures into bone and
cartilage matrices during growth and remodeling. Differentiation and remod-
eling of bone and cartilage varied not only with age, but also between regions
of attachment of single muscles and ligaments. Insertion regions were charac-
terized by the presence of coarse-fibered periosteal bone and chondroid bone,
both morphologically intermediate between fibrocartilage and lamellar bone.
These results provide evidence that periosteal attachments, characterizing the
soft-tissue bone interface, are a necessary structural prerequisite for compen-
satory movement and invariance of the relative positions of muscles, tendons,
and ligaments during long bone growth.

During normal long bone growth in length,
muscles, tendons, and ligaments apparently
maintain the same relative positions, both
with respect to each other and with respect
to the ends of long bones (Grant et al., '78,
'80). Compensatory movement of soft struc-
tures occurs during growth; thus muscles at-
tached to the metaphysis of a long bone are
not "left behind" near or at the middle of the
diaphysis following increases in bone length.
Grant and co-workers conducted experi-
ments designed to identify factors contribut-
ing to the regulation of muscle translocation
during long bone growth. Muscle function
was modified by: 1) repositioning tendons
proximal and distal to their normal attach-
ments (Grant and Hawes, '77; Grant, '78)
and 2) paralysis induced by nerve or muscle
section (Grant et al., '81). When the posi-
tional relationships of experimental hind-
limb muscles were compared to contralateral
control muscles, the two sides were not sig-
nificantly different. Thus, it was concluded
that changing either the magnitude or direc-
tion of muscle force did not affect the final
positions of the attached structures; migra-
tion was not dependent on the biological
function of the attached structures.
When growth in length of the tibia and
fibula was disturbed in rabbits by proximal
or distal epiphysiodesis (Grant et al., %l),
significant differences between control and
experimental hindlimb muscle positions did
not occur. Compensatory growth of the intact
Address reprint requests to Jack R. Hurov, PhD, Department
of Orthodontics, Washington University School of Dental Medi-
cine, 4559 Scott Avenue, St. Louis, MO 63110.
314 J.R. HUROV

epiphysis was sufficient to maintain the po- fied in 5% formic acid for up to 4 days. Tissue
sitional relationships of soft structures at- blocks were dehydrated in graded ethanol,
tached to the leg. Of importance, however, cleared, oriented, and embedded in paraffin.
was the fact that growth a t the proximal or Serial sections were cut transversely (popli-
distal end of the tibia was reduced (70% and teus and semitendinosus muscles and exten-
50%, respectively), rather than totally abol- sor retinaculum) and longitudinally (poplite-
ished following epiphysiodesis. us muscle and medial collateral knee liga-
Grant and co-workersreintroduced a model ment) at 7 pm and stained with Verhoeff’s
which could account for their observations of elastic tissue stain, Alcian-blue and periodic
muscle “migration”-the elastic sleeve model acid-Schiff for acid mucopolysaccharides (Lev
of the periosteum (Grant et al., ’80). This and Spicer, ’64), hematoxylin and eosin
model invoked interstitial periosteal growth (H&E), and with aldehyde fuchsin counter-
and diaphyseal elongation to explain both stained both with Weigert’s iron hematoxy-
compensatory movement and invariance of lin and Van Gieson’s picro-fuchsin.
the relative positions of soft structures at- Addison (‘73)recommended aldehyde fuch-
tached to long bones during ontogeny. sin counterstained with Van Gieson’s picro-
Periosteal expansion, together with dia- fuchsin to demonstrate cartilage and bone
physeal growth in length, had been recog- vividly and simultaneously. The characteris-
nized for more than a century to be the most tic purple of aldehyde fuchsin was evident in
reasonable mechanism to account for the mi- cartilage matrix and elastic fibers, while
gration of attached soft structures (Hum- bone matrix was stained red by picro-fuchsin.
phry, 1861;Warwick and Wiles, ’34;Lacroix, In this study, Weigert’s iron hematoxylin was
’51; Dorfl, ’80a,b). However, Grant and co- incorporated into Addison’s (‘73)protocol as
workers (‘80) initially formalized the elastic an additional stain, resulting in enhanced
sleeve model based on experimental studies. cellular detail.
Moreover, they proposed three testable hy- Stained sections were photographed
potheses consistent with the model: 1) soft through a Zeiss standard light microscope.
structures are attached to perwsteum rather Measurements of tissue thicknesses were
than to subjacent bone during growth; 2) the made directly from slides with the aid of an
periosteum is attached solely to the epi- ocular micrometer. Because apparent tissue
physes of long bones and must be capable of thicknesses depend on the plane of section,
sliding along the diaphysis; and 3) periosteal reported measurements are directly compa-
expansion is equal at all levels. rable within individual sections only.
A detailed description of the normal mor- RESULTS
phology of the soft tissue-bone interface is Gross anatomy
necessary to understand how soft structures
maintain their positional relationships dur- The popliteus muscle possessed a broad
ing growth. The purpose of the present study quadrilateral attachment on the posterior
is to determine the manner in which soft surface of the proximal tibial diaphysis. Mus-
structures-periosteum, epimysium, tendons, cle fibers converged onto a round tendon that
and ligaments-attach to long bones in grow- inserted into a shallow depression inferior to
ing rabbits. the lateral epicondyle of the femur.
The semitendinosus muscle arose in part
MATERIALS AND METHODS from the ischial tuberosity and in part from
New Zealand white rabbits (Oryctolagus the fascia of the biceps femoris muscle. In the
cuniculus), aged 2, 21,49,and 60 days (two thigh, muscle fibers converged onto a flat
at each age), were used for histologic exami- tendon that inserted into the anteromedial
nation. Left and right knees and distal tib- surface of the proximal tibial diaphysis.
iae, together with the popliteus (diffuse, The medial collateral ligament of the knee
diaphyseal attachment) and semitendinosus attached to the medial epicondyle of the fe-
muscles (restricted, diaphyseal attachment), mur. Fibers coursed inferiorly and somewhat
collateral knee ligaments (restricted diaphy- anteriorly, inserting on both the medial tib-
seal-epiphyseal attachments), and extensor ial condyle and posteromedial surface of the
retinacula (restricted diaphyseal attach- tibial diaphysis.
ments), were freed by dissection from anes- The extensor retinaculum (= anterior an-
thetized rabbits. nular or crural ligament) attached to the an-
Tissue blocks were fixed by immersion in teromedial surface of the distal tibial
10% neutral buffered formalin and decalci- diaphysis. Fibers coursed inferolaterally, in-
 AGE CHANGES OF SOFT-TISSUEBONE INTERFACE 3 15

Fig. 1. Popliteus muscle insertion (aldehyde fuchsin).
A) Forty-nineday-oldrabbit, longitudinal section. Arrow
indicates transperiosteal collagen fibers contacting la-
mellar bone. B)Sixty-day-oldrabbit, transverse section.
Arrows indicate periosteal fibers embedded in resorptive
bone surface. Arrowhead indicates adjacent resorptive
surface lacking periosteal fiber connections. EF, elastic
fibers; EP, epimysium; LB, lamellar bone; M p , muscle
fibers of popliteus muscle; P, periosteum.

serting on the anterior surface of the distal periosteal collagen fibers, ranging in thick-
fibular diaphysis. ness from 1-5 pm, could be traced to their
terminations on the periosteal surface of la-
Histology mellar bone (Fig. 1A). Aldehyde fuchsin
General observations clearly demonstrated periosteal elastic fibers
A clear morphologic distinction between fi- arranged in layers (Fig. 1B).
brous and cellular periosteum (or perichon- In rabbits aged 60 days, compact lamellar
drium) was often lacking histologically and bone predominated in all cross-sections ex-
division was arbitrary. This was particularly amined (Fig. 1B). The periosteal surface of
evident in rabbits younger than 60 days of the tibia1 diaphysis was characterized by
age; both the fibrous periosteum and peri- shallow concavities and pale staining bone
chondrium were packed with cells. Similarly, matrix suggestive of bone resorption. None-
the subjacent cellular layers of perichon- theless, transperiosteal collagen fibers re-
drium and periosteum possessed a frame- mained intact and extended up to 25 pm
work of collagen and elastic fibers. within surface irregularities before inserting
into pale-staining bone matrix. Adjacent re-
M. popliteus sorptive diaphyseal surfaces lacked such fi-
At all ages examined, the characteristic brous connections.
cross-banding of multinucleated skeletal
muscle cells was visible. Epimysium of the Medial collateral ligament
popliteus muscle inserted into periosteum, In 2-day-old rabbits, the femoral portion of
ranging in thickness from 8 4 5 pm. Trans- the medial collateral ligament inserted into
316 J.R. HUROV

Fig. 2A-C. Medial collateral ligament insertion, longitudinal sections (aldehyde fuchsin).A)
Femoral attachment, 2-day-old rabbit. B) Tibia1 attachment, 2-day-old rabbit. C ) Femoral
attachment, 21day-old rabbit. FC, fibrocartilage;HC, hyaline cartilage; L, medial collateral
ligament; P, periosteum; P*, perichondrium.
 Fig. 2D-H. Medial collateral ligament insertion, lon-
gitudinal sections, 49-day-old rabbit. D) Femoral attach-
ment (Alcian-blue PAS). Arrowheads indicate matrix
spicules of hyaline cartilage. E) Femoral attachment (al-
dehyde fuchsin). CB, chondroid bone; FCi, inner zone of
fibrocartilage; FCo, outer zone of fibrocartilage; L, me-
dial collateral ligament; LB, lamellar bone; P, perichon-
drium; VC, vascular canal. F) Tibial attachment
(aIdehyde fuchsin). Arrowheads indicate oblique colla-
gen fiber bundles inserting into coarse-fiberedperiosteal
bone. L, medial collateral ligament; LB, lamellar bone;
P, periosteum; PB, coarse-fiberedperiosteal bone. G, H)
Medial collateral ligament insertion, longitudinal sec-
tions, 60day-old rabbit. G ) Femoral attachment (alde-
hyde fuchsin). H) Tibial attachment (H&E). CB,
chondroid bone; L, medial collateral ligament; LB, la-
mellar bone; P, periosteum.
318 J.R. HUROV
perichondrium (Fig. 2A). Condylar cartilage
accumulated an extracellular matrix rich in
acid mucopolysaccharides as evidenced by in-
tense aldehyde fuchsin staining. Peripheral
chondrocyte aggregates, toward the liga-
ment-perichondrium interface, were sepa-
rated by less intensely stained matrix sug-
gesting a relative decrease in acid mucopo-
lysaccharide concentration.
The tibial portion of the medial collateral
ligament inserted into periosteum; at the lig-
ament-periosteum interface, the periosteum
was approximately 75 pm thick (Fig. 2B).
In 21-day-old rabbits, the femoral portion
of the medial collateral ligament penetrated
epiphyseal perichondrium and inserted into
a layer of fibrocartilage, ranging in thickness
from 150-175 pm (Fig. 2C). Fibrocartilage,
continuous with subjacent hyaline cartilage,
was characterized by coarse-textured (3-13
pm thick) parallel fiber bundles, arranged in
a penniform or basket-weave manner. Encap-
sulated chondrocytes were organized in par-
allel rows between adjacent collagen fiber
bundles. Both the non-collagen matrix and
pericellular capsules accumulated aldehyde
fuchsin indicative of acid mucopolysac-
charides.
The mode of attachment of the medial col-
lateral ligament to the tibia did not differ
significantly from the condition observed in
2-day-old rabbits-the ligament inserted into
fibrous periosteum approximately 80 pm
thick.
In 49-day-old rabbits, the femoral portion
of the medial collateral ligament blended
with fibrous perichondrium and penetrated
fibrocartilage ranging in thickness from 200-
300 pm (Fig. 2D,E). The outermost fibrocar-
tilage zone, continuous with the inserting
ligament, was structurally indistinguishable
from that observed in 21-day-old rabbits.
However, an inner zone was also observed in
which chondrocyte capsules stained in-
tensely for acid mucopolysaccharides, rela-
tive to cells in the outer zone.
Chondroid bone was observed subjacent to
the inner fibrocartilage zone (Fig. 2D,E).
Chondroid bone was intermediate morpho-
logically between fibrocartilage and lamellar
bone and was characterized by: 1) a patchy
distribution of matrix stained for acid muco-
polysaccharides; 2) cells similar to chondro-
cytes in both size and shape, but irregular in
arrangement, relative to the parallel arrays
characteristic of fibrocartilage; 3) epiphyseal
cartilage matrix spicules; 4) collagen fiber
bundles, ranging in thickness from 0.25-15
pm, organized in both parallel arrays and in
a woven manner; and 5) abundant vascular
channels.
The tibial portion of the medial collateral
ligament maintained its periosteal attach-
ment as observed in 21-day-old rabbits (Fig.
2F). Unlike the condition observed in younger
rabbits, however, coarse-fibered periosteal
bone had differentiated subjacent to the at-
tachment of the ligament. Parallel collagen
fiber bundles, extending obliquely through
cellular periosteum, were continuous with
periosteal bone. Presumably, these fibers
united the ligament, by way of the fibrous
periosteum, to subjacent bone.
Coarse-fibered periosteal bone, like chon-
droid bone, was intermediate morphologi-
cally between fibrocartilage and lamellar
bone. Coarse-fibered periosteal bone was
characterized by an irregular lattice of colla-
gen fiber bundles. The long axes of cells were
aligned parallel to major fiber axes within
limited territories. In contrast to fibrocarti-
lage, vascular channels were commonly ob-
served. Moreover, neither the matrix nor cell
boundaries accumulated stains specific for
acid mucopolysaccharides. Relative to lamel-
lar bone, the matrix of coarse-fibered perios-
teal bone was stained heterogeneously by
Van Gieson’s picro-fuchsin, presumably re-
flecting variable collagen fiber densities and
degree of mineralization of the extracellular
matrix.
In 60-day-old rabbits, the medial collateral
ligament inserted into femoral and tibial
periosteum. Fibro- and hyaline cartilage ob-
served at the femoral attachment of the lig-
ament in rabbits less than 60 days of age,
had been replaced by chondroid and lamellar
bone (Fig. 2G). Moreover, coarse-fibered per-
iosteal bone, characteristic of the tibial at-
tachment in rabbits less than 60 days of age,
had been remodeled into compact lamellar
bone (Fig. 2H).
M. semitendinosus
In 2-day-old rabbits, parallel fibers of the
semitendinosus tendon inserted into tibial
periosteum. Fibrous periosteum ranged from
5-15 pm in thickness, while the cellular peri-
osteum ranged from 8-20 pm in thickness.
In rabbits older than 2 days of age, the
semitendinosus tendon inserted into a local
thickening of periosteum on the anterome-
dial surface of the proximal tibia (Fig. 3A).
The tendon-periosteum interface averaged
 AGE CHANGES OF SOFT-TISSUE BONE INTERFACE 319

Fig. 3. Semitendinosus tendon insertion, transverse sections, (aldehyde fuchsin). A) Twenty-

one-day-old rabbit. Asterisk indicates endochondral bone; arrows indicate intramembranous
bone. B) Sixty-day-oldrabbit. Arrowheads indicate strip of chondroid bone interposed between
cartilage and lamellar bone. CB, chondroid bone; FC, fibrocartilage; FP, fibrous periosteum;
HC, hyaline cartilage; LB, lamellar bone; T, semitendinosus tendon.

0.75 mm in thickness and was characterized At 2 days of age, the tibial cortex was com-
by parallel and oblique arrangements of col-
posed of primary spongiosa, which extended
around the cartilaginous fibula as a peri-
lagen fibers. Toward the tibial cortex, colla-
gen fibers were arranged longitudinally. chondrial collar of intramembranous bone
Medullary cavities of the tibiae in both 21-
(Fig. 4A,B). By 21 days of age, however, cir-
and 49day-old rabbits were composed of tra-cumferential primary spongiosa had been re-
beculae formed by endochondral ossification;
sorbed subperiosteally opposite the tibial
osteoid had been deposited on cartilage ma-attachment of the retinaculum, thereby ex-
trix spicules and hypertrophied chondrocytes
posing trabeculae of endochondral bone (Fig.
(Fig. 3A). However, tibial cortices were com-
4C).
posed of primary spongiosa characteristic of In contrast to 21-day-oldrabbits, 49-day-old
intramembranous ossification. rabbits revealed coarse-fibered periosteal
In 60-day-old rabbits, both cartilage andbone opposite the tibial attachment of the
chondroid bone had differentiated subperios-
retinaculum (Fig. 4E). Transperiosteal colla-
teally, forming integral parts of the insertion
gen fiber bundles, presumably continuous
of the semitendinosus tendon (Fig. 3B). with the retinaculum, could be traced to their
insertions into coarse-fibered periosteal bone
Extensor retinaculum (Fig. 4G).
In rabbits aged 2, 21, and 49 days, the By 60 days of age, coarse-fibered periosteal
extensor retinaculum inserted into fibrous bone and primary spongiosa had been re-
periosteum, from 70-100 pm thick, on both placed by compact lamellar bone. At the tib-
the tibia and fibula (Fig. 4A-F). ial attachment of the extensor retinaculum,
320 J.R. HUROV

Fig. 4A-D. Extensor retinaculum, transverse sections.
A,B) Tibial (A) and fibular 03) attachment in 2-day-old
rabbit (aldehyde fuchsin). C,D) Tibial (C) and fibular @)
attachment in 21-day-old rabbit (Alcian-bluePAS). Note
absence of cortical primary spongiosa subjacent to tibial
attachment of retinaculum in 4C. Arrowheads in 4D
indicate cortical primary spongiosa. HC, hyaline carti-
lage; FP, fibrous periosteum; P, periosteum;PS, primary
spongiosa; R, extensor retinaculum.
 AGE CHANGES OF SOFT-TISSUEBONE INTERFACE 321

Fig. 4E-G. Extensor retinaculum, transverse sec- cate demarcation between coarse-fiberedperiosteal bone
tions, 49day-old rabbit (aldehyde fuchsin). Tibia1 attach- and lamellar bone. F) Fibular attachment. CP, cellular
ment; rectangular area in 4E enlarged in 4G. Small periosteum; FP, fibrous periosteum; LB, lamellar bone;
. _ in G-.
arrows indicate collagen
~ . . .fiber bundles continuous
. .. PB, coarse-fibered periosteal bone; R, extensor reti-
322 J.R. HUROV

Fig. 4H, I. Extensor retinaculum, transverse sections, 60-day-old rabbit. H) Tibia1 attach-
ment (aldehyde fuchsin). Arrowheads indicate fiber tracts within lamellar bone. I) Fibular
attachment (Verhoeffs). Periosteum is absent; retinaculum inserts directly into cortical lamel-
lae (stained black by Verhoeff s). EF, elastic fibers; LB, lamellar bone; HS, Haversian system;
R, extensor retinaculum.
 AGE CHANGES OF SOFT-TISSUE BONE INTERFACE
TABLE 1. Attachments of soft structures in rabbits aged 2-60 davs
Structure
(attachment site) 2 21
M. popliteus Periosteum
(diffuse, diaphyseal)
M. semitendinosus Periosteum
(restricted, diaphyseal)
Medial collateral ligament
Femur Perichondrium
(restricted, epiphyseal)
Tibia Periosteum
(restricted, diaphyseal)
Extensor retinaculum
Tibia Periosteum
(restricted, diaphyseal)
Fibula Periosteum
(restricted, diaphyseal)
collagen fiber bundles penetrated periosteum
and entered circumferential lamellar bone;
fiber tracts extended 120-380 pm into bone
(Fig. 4H). Randomly oriented periosteal elas-
tic fibers were abundant at the soft tissue-
bone interface. The fibular portion of the ex-
tensor retinaculum inserted directly into cir-
cumferential lamellar bone, and periosteum
was absent at this site only (Fig. 4I).
The manner in which soft structures in-
serted and the types of bone and cartilage
encountered at soft tissue-bone interfaces,
as a function of growth, are summarized in
Table 1.
DISCUSSION
Results of the present study confirmed that
soft structures attached principally to fibrous
periosteum or perichondrium from 2-60 days
of age. The presence of an extensive soft tis-
sue-perwsteal interface in growing rabbits is
consistent with the hypothesis proposed by
Grant et al. (‘80) concerning the mode of at-
tachment of soft structures, and accords with
observations made by Lacroix (‘511, Bier-
mann (‘57), and Dorfl(‘80a).
Grant and co-workers (‘80) hypothesized
that the periosteum was attached solely t o
1the epiphyses of long bones; the present his-
tologic results do not support this hypothesis.
A collagen framework within cellular peri-
osteum, established at least by 2 days of age,
linked fibrous periosteum and perichon-
drium to subjacent bone or cartilage matri-
ces. Fibers within the cellular periosteum
323
Mode of attachment bv age (davs)
49 60
Periosteum Periosteum Periosteum
Periosteum Periosteum Periosteum,
fibrocartilage,
hyaline cartilage,
chondroid bone
Perichondrium, Perichondrium, Periosteum,
fibrocartilage fibrocartilage, chondroid bone,
chondroid bone lamellar bone
Periosteum Periosteum, Periosteum,
coarse-fibered lamellar bone
periosteal bone
Periosteum Periosteum, Periosteum,
coarse-fibered lamellar bone
periosteal bone
Periosteum Periosteum Lamellar bone
have been referred to as the “second relay of
collagenous fibers” by Moss and Moss-Salen-
tijn (‘78, p. 47). Oblique fiber bundles linking
fibrous periosteum to subjacent bone were
similarly observed in human femora and hu-
meri by Biermann (‘57), rat humeri by Hark-
ness and Trotter (‘78), and in rabbit tibiae by
Dorfl (‘80a). Collagen fibers of epimysium,
tendons, and ligaments thus comprise a sys-
tem both structurally and functionally inte-
grated with collagen fibers of periosteum and
subjacent bone and cartilage matrices.
Grant et al. (‘81) argued that epiphyseal
elongation stretches the periosteum, thus
passively carrying attached soft structures
during growth. Their suggestion gains sup-
port from experimental work; when tibial
elongation was completely arrested in grow-
ing rabbits by epiphyseal stapling (Dorfl,
’80b), periosteal expansion and migration of
tendons and ligaments were halted. Soft
structures anchored to fibrous periosteum,
therefore, “hitchhiked” during longitudinal
bone growth and periosteal expansion, thus
maintaining constant positional relation-
ships. This mode of soft tissue translocation
is significant because the rabbit tibia attains
about 60% of its adult length .between 2-60
days of age (Heikel, ’60). The presence of an
extensive collagen framework linking fi-
brous periosteum or perichondrium to subja-
cent bone and cartilage matrices precludes
periosteallperichondrial sliding directly along
the tibial diaphysis, however. Further re-
search is needed to determine both perios-
324 J.R. HUROV
teallperichondrial growth sites and elon-
gation planes.
Periosteal elongation, in response to ten-
sile loads engendered during epiphyseal car-
tilage proliferation, presumably stimulates
local periosteal growth. Yeh and W a n (‘84)
showed that mechanical stretching of colla-
gen ribbons in vitro, on which fetal rat cal-
varial cells were grown, promoted significant
increases in prostaglandin synthesis; Raisz
and Martin (‘83) proposed that prostaglandin
synthesis regulated bone metabolism. It is
tempting to speculate that periosteal tension
stimulates growth by a similar mechanism.
Moreover, periosteal elastic fibers may play
a role in force transmission, in vivo.
The present histologic study showed that
differentiation and remodeling of cartilage
and bone at soft tissue-bone interfaces varied
not only with age, but also between attach-
ment regions of single muscles and liga-
ments according to: 1)attachment type, i.e.,
diffuse (fleshy) or restricted (tendon or liga-
ment); and 2) attachment site, i.e., epiphysis
or diaphysis (Table 1).These data support
Hoyte and Enlow’s (‘66) suggestion that bone
differentiation and remodeling must be lo-
cally regulated. Since the periosteum repre-
sents a major source of osteoprogenitor cells
(Tonna and Cronkite, ’63), it follows that the
periosteum must also be locally regulated.
Temporal patterning of: 1)appearance, and
2) remodeling of cartilage and bone (Table 1)
suggests that maturation of soft tissue-bone
interfaces may be viewed from two related
perspectives. The first stresses cartilage and
bone differentiation. The second emphasizes
incorporation of soft structure attachments
into bone and cartilage matrices resulting
from growth and remodeling.
The presence of chondroid bone, interposed
between fibrocartilage and lamellar bone at
the femoral attachment of the medial collat-
eral ligament, suggested fibrocartilage was
being converted directly to bone (Fig. 2D,E).
Presumably, chondroyctes were responsible
for: 1) altering the relative proportions of
extracellular matrix components, and 2) me-
diating ossification. Using immunofluo-
rescence techniques with collagen specific
antibodies, von der Mark and von der Mark
(‘77) showed that hypertrophic chondrocytes
in chick tibial epiphyses stained for type I
collagen. This finding suggested that endo-
chondral osteoid was initially synthesized,
and presumably secreted, by chondrocytes.
Whether or not fibrocartilage cells possess
osteogenic potential, in vivo, remains contro-
versial (Videman, ’70; Beresford, ’81, p. 96),
but is a subject inviting further research.
Results of this histologic study showed that
fibrocartilage and coarse-fibered periosteal
bone were similar morphologically. Collagen
fibers were arranged in a parallel or penni-
form manner. The major axes of cells and
collagen fibers were oriented in parallel.
Moreover, fibrocartilage contains types I and
11 collagen (Boyde and Jones, ’83). Coarse-
fibered periosteal bone, characteristic of the
tibial attachments of the medical collateral
ligament and extensor retinaculum (Table l),
differentiated within diaphyseal periosteum,
rather than from fibrocartilage. Petersen (‘30)
and Biermann (‘57) termed periosteally de-
rived coarse-fibered bone “Einstrahlungs-
knochen.”
The simultaneous presence of cartilage,
chondroid bone, and lamellar bone and, sim-
ilarly, coarse-fibered periosteal bone, to-
gether with lamellar bone at tendon and
ligament insertions, is not surprising (Table
1).These tissues had been repeatedly co-dem-
onstrated in a variety of mammalian species
by Weidenreich (‘23a,b, ’301, Dolgo-Saburoff
(‘29), Petersen (‘30), Cooper and Misol (‘70),
and Becker (‘71). Temporal patterns of cell
differentiation and modulation of extracellu-
lar matrix products depend, in part, on epi-
genetic factors, such as prevalent mechanical
conditions(Weidenreich, ’23b, p. 596) and cell
position (Knese and Biermann, ’58,pp. 179-
182). Growth and remodeling will also dic-
tate spatial organization and volumes of co-
existent cartilage and bone matrices. A strict
deterministic sequence of cartilage and bone
appearance, as a function of maturation of
soft tissue attachments, is therefore pre-
cluded.
To better understand the roles of long bone
growth and periosteal expansion for main-
taining the relative positions of attached soft
structures, the following questions should be
addressed. Is the timing of cartilage and bone
differentiation at soft tissue-bone interfaces
matched with long bone and periosteal
growth rates? Do cartilage and bone differ-
entiation and remodeling coincide with de-
creased epiphyseal and periosteal growth
rates? Would premature arrest of epiphyseal
growth result in the accelerated appearance
of cartilage and bone at soft tissue-bone
interfaces?
ACKNOWLEDGMENTS
I thank Pere Alberch, James Yaeger, and
two anonymous readers for careful reviews
 AGE CHANGES OF SOFT-TISSUE BONE INTERFACE
and suggestions for improving this paper.
April Rahing’s expert technical assistance in
preparing the photomicrographs is gratefully
acknowledged. The histology portion of this
study was done while the author was a re-
search assistant with Philip Grant at The
University of Texas In Austin. Research was
supported by grants PHS AM19632 to Philip
Grant and NIDR 5T32 DE07047 to the De-
partment of Orthodontics, University of Con-
necticut Health Center, Farmington.
LITERATURE CITED
Addison, P.C. (1973) Advantages in the use of aldehyde
fuchsin with Van Gieson’s picro-fuchsin counterstain-
ing for differentiating bone and cartilage. Stain Tech-
nol. 48:59-62.
Becker, W. (1971) Elektronenmikroskopische Untersu-
chung der Insertion van Sehnen am Knochen. Arch.
Orthop. Unfall-Chir. 69:315-329.
Beresford, W.A. (1981) Chondroid Bone, Secondary Car-
tilage and Metaplasia. Baltimore: Urban and
Schwarzenberg.
Biermann, H. (1957) Die Knochenbildung im Bereich
periostaler-diaphysarer Sehnen und Bandansatze. Z.
Zellforsch. 46:675-671.
Boyde, A., and S. Jones (1983) Scanning electron micros-
copy of cartilage. In B.K. Hall (ed): Cartilage Struc-
ture, Function, and Biochemistry, Val. 1. New York
Academic Press, pp. 105-148.
Cooper, R.R., and S. Misol (1970) Tendon and ligament
insertion: A light and electron microscope study. J.
Bone Joint Surg. 52A:1-21.
Dolgo-Saburoff, B. (1929) Uber Ursprung und Insertion
der Skelettmuskeln. Anat. Anz. 68:80-87.
Dorfl, J. (1980a) Migration of tendinous insertions. I.
Cause and mechanism. J. Anat. 131:179-195.
Dorfl, J. (1980b) Migration of tendinous insertions. II.
Experimental modifications. J. Anat. 131:229-237.
Grant, P.G. (1978) The effect of position on the migration
of muscle. J. Anat. 127:157-162.
Grant, P.G., and M. Hawes (1977) Experimental modifi-
cation of muscle migration in the rabbit. J. Anat.
123:361-367.
Grant, P.G., P.H. Buschana, and D.W. Drolet (1978) Po-
sitional relationships of-structures attached to long
bones during growth Cross-sectional studies. Acta
Anat. 102378-384.
Grant, P.G., P.H. Buschang, D.W. Drolet, and C. Picker-
ell (1980) Invariance of the relative positions of struc-
tures attached to long bones during growth: Cross-
sectional and longitudinal studies. Acta Anat. 107:26-
34.
Grant, P.G., P.H. Buschang, D.W. Drolet, and C. Picker-
ell (1981)The effect of changes in muscle function and
bone growth on muscle migration. Am. J. Phys. An-
thropol. 54:547-553.
Harkness, E.M., and W.D. Trotter (1978) Growth of
transplants of rat humerus following circumferential
division of the periosteum. J. Anat. 126275-289.
Heikel, H.V.A. (1960) On ossification and growth of cer-
325
tain bones of the rabbit; with comparison of the skele-
tal age in the rabbit and man. Acta Orthop. Scand.
29:171-184.
Hoyte, D.A.N., and D.H. Enlow (1966) Wolffs Law and
the problem of muscle attachment on resorptive sur-
faces of bone. Am. J. Phys. Anthropol. 24:205-214.
Humphry, G.M. (1861) Observations of the growth of the
long bones and of stumps. Med-Chir. Trans. 44:117-
134.
Knese, K-H., and H. Biermann (1958) Die Knochenbil-
dung an Sehnen- und Bandansatzen im Bereich ur-
spriinglich chondraler Apophysen. Z. Zellforsch.
49:142-187.
Lacroix, P. (1951) The Organization of Bones. London:
Churchill, pp. 58-77.
Lev, R., and S.S. Spicer (1964) Specific staining of sul-
phate groups with Alcian blue at low pH. J. Histo-
chem. Cytochem. 12309.
Moss, M.L., and L. Moss-Salentijn (1978) The muscle-
bone interface. In D.S. Carlson and J.A. McNamara,
Jr. (eds): Muscle Adaptation in the Craniofacial Re-
gion. Monogr. 8, Craniofacial Growth Series, Center
for Human Growth and Development. Ann Arbor: Uni-
versity of Michigan, pp. 39-71.
Petersen, H. (1930) Die Organe des Skelettsystem. In W.
v. Mollendorf (ed): Handbuch der mikroskopischen An-
atomic des Menschen (Stutzgewebe, Knochengewebe,
Skelettsystem). Berlin: Springer, pp. 521-678.
Raisz, L.G., and T.J. Martin (1983) Prostaglandins in
bone and mineral metabolism. In W.A. Peck (ed): Bone
and Mineral Research Annual II. Amsterdam: Ex-
cerpta Medica, pp. 286-310.
Tonna, E.A., and E.P. Cronkite (1963) The periosteum:
Autoradiographic studies on cellular proliferation and
transformation utilizing tritiated thymidine. Clin. Or-
thop. 30:218-232.
Videman, T. (1970) An experimental study of the effects
of growth on the relatiokhip of tendonsand ligaments
to bone at the site of diaphyseal insertion. Part I E
Autoradioerauhic studv with 3H-thvmidine of the in-
sertion area bf the &don of the “pectineus muscle.
Ann. Chir. Gynaecol. 59:35-41.
von der Mark, K., and H. von der Mark (1977) The role
of three genetically distinct collagen types in endo-
chondral ossification and calcification of cartilage. J.
Bone Joint Surg. 59B3458-464.
Wanvick, W.T., and P. Wiles (1934) The growth of peri-
osteum in long bones. Br. J. Surg. 22169-174.
Weidenreich, F. (1923a) Knochenstudien. I. Teil. f h e r
Aufbau und Entwicklung des Knochens und den Char-
akter des Knochengewebes. Z. Anat. Entwickl. (Abt. 1)
69~382466.
Weidenreich, F. (1923b) Knochenstudien. II. Teil. %r
Sehnenverknocherungen und Faktoren der Knochen-
bildung. Z. Anat. Entwickl. (Abt. 1)69:558-597.
Weidenreich, F. (1930) Das Knochengewebe. In W. v.
Mollendorf (ed): Handbuch der mikroskopischen Ana-
tomie des Menschen (Stutzgewebe, Knochengewebe,
Skelettsystem). Berlin: Springer, pp. 391-520.
Yeh, C-K., and G. Rodan (1984) Tensile forces enhance
prostaglandin E synthesis in osteoblastic cells grown
on collagen ribbons. Calcif. Tissue Int. 36:S67-S71.