вход по аккаунту


An examination of alternative explanations of split-line orientation in compact bone.

код для вставкиСкачать
An Examination of Alternative Explanations of
Split-line Orientation in Compact Bone '
Division of Orthopaedics and Department of Sociology and Anthropology,
Tulane University, N e w Orleans, Louisiana
The split-line technique reveals structural orientation in many areas of the
skeleton in a variety of mammals (Benninghoff, '25; Bruhnke, '29; Henckel, '31;
Tappen, '53). Certain regions of the facial
skeleton in primates showed consistent
patterns which appear to be related to mechanical forces acting upon the bone (Tappen, '53, '54, '57, '63). This viewpoint has
been disputed on theoretical grounds by
Moss ('54), Evans and Goff ('57) and
Evans ('57). Experimental verification was
also lacking.
Subsequent experiments by Avis ('59)
and Holmstrand ('57) utilizing surgical
interference are relevant to the problem,
but the results appear to be conflicting.
There are also technical deficiencies in
both studies, and practical difficulties in interpreting their results.
It is possible that more than one cause
is operating to bring about observable splitline organization, or that different processes are acting in different parts of the
skeleton. Accordingly, explanations other
than that of response to mechanical forces
also need to be examined. Of these, only
two have been offered, to my knowledge.
Both have been set forth by Evans and
Goff ('57) in their discussion of other
studies of bone organization.
The first hypothesis is that the splitlines are sampling the orientation of the
bone structure in the direction of growth
of the bone. Thus the split-lines made in
long bones generally course in the same
direction as the axis of the shaft, which is
the main direction of longitudinal growth.
Similarly, Ahrens ('36) found that splitlines in the roofing bones of the skulls of
human infants generally follow the pattern of bone growth radiating from the
centers of ossification.
AM. J. PHYS. ANTHROP.,22: 423-442.
The other hypothesis is that the splitlines are following the course of the vascular pattern within the bone, the canals
which carry the blood supply to the bone
tissue. These canals are very numerous in
bone, and in general are seemingly oriented parallel to the axis of long bones.
The present paper offers evidence which
tends to contradict the explanations set
forth by Evans and Goff, and discusses
some of the criticisms of the "mechanical
forces" hypothesis. This is part of a larger
study of the microscopic structural basis
for the presence or absence of split-line orientation in different species and in different parts of the skeleton.
The split-line technique has been described in detail elsewhere (Seipel, '48;
Tappen, '54). Partially decalcified compact bone may be readily punctured by a
needle. This causes short crevices to form
in many regions; in some areas only round
or irregular holes result. Ink inserted into
the splits makes them readily visible, providing a permanent record. In areas where
the bone has a strong orientation parallel
to the surface, splits may be joined to
form continuous lines, or the point of
a teasing needle inserted to varying
depths may be pulled continuously in the
direction of least resistance, giving a similar but somewhat smoother and more precise result. The latter procedure i s most
commonly used in this laboratory.
Split-line preparations on different surfaces of the same bone often have the
same major orientations. However, several bones processed recently in this lab1 Supported in part by N. I. H. grant HD-00510.
The technical assistance of Orlando Fernandez, Aura
Maldonado, Carolyn Pons and Jayne Stouse is gratefully acknowledged.
oratory show marked differences in splitline direction along opposite surfaces : two
human innominates, one dog scapula, and
two human scapulae. Examples are shown
in figures 1-4, and they are described in the
“Results” section. Other preparations showing divergent patterns on opposed surfaces
which have been reported in other publications are also referred to in the discussion. These preparations give a partial test
of the hypothesis of direction of growth of
a bone as the determinant of its split-line
Split-lines can also be examined microscopically. Methods for block staining and
gelatin embedding decalcified bone devised in this laboratory (Tappen, ’62)
greatly facilitate the preparation of serial
sections for microscopic study. It is easy
to follow split-lines serially, since the gelatin embedding medium is tough and flexible enough to maintain the tissue relationships, without tearing or folding, after
the fissures are produced by the needle.
Striking features of the vascular canals
revealed by serial studies of cross sections
of the middle part of the shaft of the
femur in three species of monkeys, Cercopithecus ascanius, Cercopithecus aethiops
and Cercocebus albigena, suggested a critical experiment for evaluating the hypothesis that orientation of the vascular canals
in bone is a basis of split-line direction.
In monkeys this part of the femur is usually composed largely of outer circumferential lamellae. In some specimens there
is a relatively narrow area near the marrow cavity made up largely of Haversian
systems. Osteones are also numerous
throughout the bone in the vicinity of the
linea aspera. Otherwise, the vascular
canals of the circumferential lamellae usually have no layers of bone surrounding
them, and the pattern of their orientation
is along the axis of the shaft but uniformly
diagonal to the axis as they penetrate gradually inward and proximally. This simple
pattern contrasts with the arrangement of
Haversian canals and Haversian systems,
which usually form a complex network of
branchings and anastomoses.
If vascular orientation determines the
direction of split-lines, this diagonal pattern of the vessels in the femoral shaft of
monkeys should be reflected in appropriate
split-line preparations in this region. Accordingly, macerated femora of two adult
males of Cercopithecus ascanius were decalcified in dilute acid and sliced coronally
so as to bisect the shaft into two parts as
nearly equal as possible. The remaining
marrow and much of the cancellous bone
were cleaned from the marrow cavity, and
split-lines were made in the cortex of the
shaft at right angles to the plane of the
cuts, as shown in figure 5a. Figure 5b
shows the variable depth of penetration of
the India ink and the plane of the coronal
section of the shaft.
The two halves of each monkey femur
were then approximated and cut into parallel-ended blocks 1 cm in length, using
a device developed in this laboratory
(Stagg and Tappen, ’ 6 3 ) for guiding the
knife-cuts. The blocks were then stained
in Harris’ hematoxylin, washed, and infiltrated with progressively thicker gelatin
solutions until the embedding strength of
45% was reached. Serial sections of each
of these blocks were cut at a thickness of
30 w, using an adaptation of a rotary microtome for cutting frozen sections. Anterior and posterior halves were mounted
separately in serial sections. Cutting and
mounting order was from proximal to distal for each block. This makes it possible
to trace structural features of the bone
serially through hundreds of sections, potentially the entire length of the bone.
Split-lines were also made in fragments
of the shafts of two femora of Cercqpithecus ascanius, some with only very slight
penetration of the needle, others nearly as
deep as the marrow cavity. Figure 6 illustrates split-line patterns in the periosteal
surface of one such femoral fragment,
showing that the presence of Haversian
systems is not essential to the production
of split-lines in bone.
Split-liTLes on opposite surfaces. Figure
1 shows split-line preparations of opposite
surfaces of the same human innominate
bone. While the split-line patterns correspond closely on opposite sides in such
areas as the iliac crest and the greater
sciatic notch, the orientation is strikingly
different on the external and internal surfaces of the greater part of the wing of the
ilium. The major orientations of the splitlines diverge a s much as 90" on the two
surfaces. A split-line preparation of a n innominate bone from another individual
confirms the deviations in orientation in
this large area. Both bones are of unknown origin. This finding agrees closely
with the split-line patterns in human ilia
reported by Mednick ('55), and the photographs show the regularity and smooth
curvatures of the split-line orientations
more clearly than do her line drawings.
Figure 2 shows split-line preparations on
costal and dorsal surfaces of the scapula
of a dog. The curvatures in split-lines of
the zone of transition near the apex of the
spine have no equivalents on the opposed
costal surface, where the split-lines continue parallel to the vertebral border. However, a double curvature does develop on
this surface farther away from the vertebral border. The recurved pattern as the
split-lines of the axillary border and the
spine converge toward the acromial process, incompletely shown in figure 2a, also
has no equivalent on the supraspinous surface, where the split-lines all continue into
the acromial process. This surface is not in
the field of view in figure 2. The transparence of the decalcified bone in thin
areas also reveals deviations in direction
on opposite surfaces, although there are
also strongly oriented major patterns
which are quite similar.
Split-line preparations on two human
scapulae of unknown origin are shown in
figures 3 and 4. Here again there are extensive areas where the opposite surfaces
coincide neatly in their split-line orientations, especially along the thickened axillary and vertebral borders. In the thinner
areas there are extensive divergences in
patterns, however. Some of these can be
detected in the photographs, because the
transparence of the decalcified bones reveals the split-lines of the opposite surface. With due caution, these divergent
splits can be made without tearing the
bone of the opposite surface, by keeping
the penetration of the needle quite shallow
as it is pulled along the surface in the
direction of least resistance. The patterns
obtained are often curved, and cannot usually be predetermined by examining the
bone surface. In the regions where the
orientation of both sides is confluent, the
resistance of the decalcified bone of the
opposite surface to the pull of the needle
is greatly reduced, and frequently results
in complete penetration and a n identity of
the split on the two surfaces.
It should be observed that the two human scapulae do not have identical splitline patterns, and particularly that the
major orientations in the supraspinous
fossa are quite different. In figure 3a the
divergence of orientation in this region
from that of the corresponding part of the
costal surface can be seen through the
bone near the superior border. In the
other scapula the patterns on these opposed surfaces are confluent.
The divergence of patterns in the human scapulae is not always in regions of
thin bone, however. The most prominent
bony ridge for tendinous intersections of
the subscapularis muscle, seen on the
costal surfaces of both scapulae, is marked
by split-lines deviating from the general
pattern and conforming to the orientation
of the ridge from the lower vertebral
border toward the lower margin of the
glenoid fossa, as seen in figures 3a and 4a.
The split-lines on the corresponding part
of the dorsal surface of the two scapulae
sweep across the ridge at virtual right
angles, as seen in figures 3b and 4b.
Microscopic tracings of vascular canals
and split-lines. Figure 7a-h shows low
power microscopic views of eight serial
cross sections from the mid-shaft region
of the femur of a monkey, prepared as in
figure 5 , and arranged serially from the
distal toward the proximal end. The splitline perpendicular to the cut surface is
visible in the central region of the cortical bone, the India ink penetrating
deeply and showing as a dark line continuous with a cement line approximately
parallel to and midway between the periosteal and endosteal surfaces. In figure 7b
there is slight overlap of the parts of the
section above and below the split-line,
especially in areas of deeper penetration.
Figure 7g shows a small gap at the periphery of the split. Osteones are absent in
the entire region.
The correspondence of the split-line to
a cement line does not seem to be accidental. Although it was manifestly impos-
sible to detect any cement lines with the
naked eye when it was being produced,
the split actually follows much of the
length of this cement line. It can be traced
through 2,563 sections, a distance of approximately 76.9 mm. In this distance,
the position of the cement line changes
only very gradually in distance from either
surface. In figure 7 it is approximately
0.60 mm from the periosteal surface and
0.77 mm from the endosteal surface. At
the proximal termination of the split-line,
the cement line is approximately 0.27 mm
from the periosteal surface and 0.82 mm
from the border of the marrow cavity. The
cement line continues proximally beyond
the end of the split. The split-line and
cement line terminate distally in an area
where they are replaced by Haversian systems associated with the linea aspera.
Here the split is approximately 0.52 mm
from the periosteal surface; the distance
from the marrow cavity cannot be determined because the angle of the cut has
not left intervening bone that can be measured perpendicularly.
By contrast with cement lines, the diagonal orientation of the great majority of
vascular canals in the series can be traced
inward by observing the changes in their
relative positions after a very few sections.
All the cement lines make convenient
markers of this progress, since their position in relation to the surfaces and to each
other remains remarkably constant. The
cement line superficial to the one carrying
the split can be traced 96.3 mm, while its
nearest neighbor on the endosteal side continues 65.8 mm, for example. The distance
between this cement line and its two
neighboring cement lines varies, in the
vicinity of the split-line, between 0.33 mm
and 0.50 mm on the periosteal side and
between 0.19 mm and 0.25 mm on the
endosteal side, during the long course the
split can be traced.
A number of the vascular canals in the
serial sequence of figure 7a-h can be
seen actually passing through cement lines
to deeper areas of the bone, including
three which pass to opposite sides of the
split-line. All the canals are noticeably
deeper at the end of the sequence, with
one exception. In that case, the canal is
joined by another, more superficial canal;
further tracing beyond the sequence shown
in the photomicrographs reveals that its
subsequent course is also inward. The
joining of the two canals may be observed
in the sequence, figure 7f, g, h, midway between the split-line and the cement line
next most superficial to it.
Figure 8a-d is a higher power sequence
from the other monkey femur processed,
showing more detail of the passage of a
canal through the split. Once more the
split conforms to a cement line. In a
sequence of 246 serial sections in this
area, 39 canals pass through this split, all
of them penetrating inward as they are
traced distally to proximally.
The distal to proximal inward orientation of the canals is typical of the course
of all lengthy non-Haversian canals in the
mid-shaft region of monkey femora, on
the basis of investigations in progress in
this laboratory. The canals can be traced
through large numbers of sections and
their orientaiton inward distally to proximally observed. For example, the canal
nearest the split-line and approximately
in the center of figure 7a passes through
the split in figure 7d and e of the sequence.
Traced superficially, this canal leaves the
periosteal surface of the bone after 90 sections. Traced toward the marrow cavity,
it leaves the bone after 147 sections. Since
sections were cut at 30 p, it has progressed
from surface to surface approximately
1.37 mm while advancing approximately
7 mm in an axial direction. Other canals
in this general area penetrate with very
similar courses, although the picture is
complicated slightly by occasional branchings, anastomoses, or blind endings of
The findings on all the split-lines made
in the cut surfaces of monkey femurs correspond closely to the situation described
here. In no areas do the split-lines correspond to the diagonal orientation of the
vascular canals. The conformity with the
orientation of the lamellae is close, however. The circumferential cement lines not
only serve as markers for this relationship;
they appear to be areas of the decalcified
bone matrix which offer less resistance to
the splitting action of the needle. The
splits are thus most likely to follow cement
lines if they are present.
A. Alternatives to “mechanical
forces” hypothesis
The evidence presented in this study indicates that neither of the two hypotheses
presented by Evans and Goff (’57) is tenable as a general explanation for the onented split-line patterns observed in decalcified bones. There is also evidence
from other sources which casts doubt upon
these hypotheses, to be discussed along
with the results reported here.
1. Direction of growth of bone as an
explanation of split-line orientation. The
photographs of opposite surfaces of the
human ilium and scapulae and the dog
scapula show radically different split-line
patterns in extensive areas. It is impossible to reconcile these opposed patterns
with the hypothesis that they are, in the
words of Evans and Goff, “directional
growth indicators.” This would require
large areas of the two surfaces of the
ilium to grow in directions as different
from each other as go”, for example.
Growth of the ilium during later stages
of development actually takes place mainly
on both tabulae near the margins of the
iliac crest, which is the site of the epiphysis capping the entire wing of the ilium
during adolescence.
While strongly opposed patterns of splitlines on opposite surfaces of the dog and
human scapulae are not as extensive as in
the human ilium, enough of them are
present to throw similar doubt upon the
notion that the split-lines are simply an
index of the direction of growth. Some
of these patterns diverge from each other
in areas so thin that the split-lines made
on the opposite surface may be clearly
seen through the bone. Thus it appears
that divergent patterns may form on opposite surfaces even when no cancellous
bone intervenes between the cortical
Photographs of other split-line preparations which may be relevant to the question of the relationship between direction
of growth and split-line orientation have
been published elsewhere. Mandibles of
several species of primates illustrated in
Tappen (’54) show quite different patterns on the lingual and labial surfaces of
the region below the incisor teeth. An
even more revealing illustration from the
above article is figure 8, showing the mandible of a spider monkey. The transparent decalcified bone allows one to see simultaneously the split-line patterns on lingual and labial surfaces of the ramus and
part of the body of the mandible. This reveals split-lines paralleling the mandibular
notch on the lateral surface while ascending with the ramus medially. There is also
a deviant pattern on the medial surface,
coursing between the angle of the jaw and
the posterior molar tooth, which may be
seen at a virtual right angle against the
sweep of split-lines on the lateral surface,
which curve to conform to the differences
in main orientation of the body and the
ramus. This deviant pattern on the lingual
surface corresponds to the area of attachment of the internal (medial) pterygoid
muscle, and it is difficult to avoid the conclusion that it is associated with the action
of this muscle.
The theory that split-lines are representations of the direction of growth of a
bone appears to be derived largely from
the pattern in long bones, where strong
orientations parallel to the axis are preponderant in the shaft. This is not the
only pattern present in long bones, however. In the case of the femur, for example, marked deviations occur in the regions
of the lesser trochanter and intertrochanteric crest, as shown in illustrations of human and other primate femora (Pauwels,
’50; Evans and Goff, ’57). On the anterior surface of the lesser trochanter the
split-lines curve away smoothly from the
axial direction of the shaft above and below the trochanter and converge toward its
summit. This pattern was noted by Evans
and Goff, but its relevance to theories of
split-line orientation was not discussed.
In view of the continuous remodeling of
the bone required to maintain the lesser
trochanter or any other proturberant structure in a relatively constant position during growth (Murray, ’36; Enlow, ’62a), it
is difficult to visualize any mechanism
which would allow major components of
growth of the base of the trochanter to be
contributed to equally and simultaneously
from above and below that structure.
In addition, the intertrochanteric crest
pattern revealed by the preparations of
Evans and Goff conforms closely to the
curvature of the crest, terminating near
the apices of the two trochanters. This pattern not only deviates from the general
axial orientation of split-lines in the shaft;
it actually interrupts the split-lines of the
shaft near the base of the lesser trochanter. Thus much the same difficulty arises
in interpreting the deviations as in the case
of the ilium, where widely diverging patterns are found on opposite surfaces.
A further complication of theories of
growth direction as an explanation for
split-line orientation arises from comparison of the patterns of the human ilium
with those in long bones. In the vicinity
of the iliac crest the split-lines are approximately parallel to the margin of the crest
on both lateral and medial surfaces. Since
much of the direction of growth of the
bone is thus at an approximate right
angle to the direction of the split-lines,
the orientational response of the bone to
the direction of growth would appear to
be very different from the response in
long bones, if split-lines are directional
growth indicators.
A study which shows clear correspondence of split-line patterns to direction of
growth in bone is that of Ahrens ('36) on
the roofing bones of the skulls of human
infants. In the newborn the split-lines radiate from the ossification centers of the
parietal and frontal bones, in the direction
of growth of the spicules of bone which
spread from these centers. This pattern
had virtually disappeared in the skull of
a child seven years of age; the splits are
random in direction on the inner surface.
This is also true of the outer surface,
though a few traces of the original arrangement remain. Thus by the age of seven
the condition of the split-lines appears to
be essentially like that reported by Benninghoff ('25) for the adult human
calotte. The original correspondence of the
split-lines and the direction of growth of
the bone has been lost, and well defined
patterns of any sort are rare. If the splitlines were primarily growth indicators,
they should be expected to remain as such
in the skull as well as in long bones.
Studies of split-line patterns in a series
of infant and juvenile gorilla skulls (Tappen, '57, '63) may also be relevant to the
problem of growth and structural orientation of compact bone. None of this series
were newborn animals, but highly oriented
split-line patterns are evident in the frontal and parietal bones of a number of specimens ranging in age from less that one
year to a possible five years.
The pattern in the frontal squama is
similar in the four youngest individuals,
the lines coursing antero-posteriorly. The
next two individuals in the age sequence
display quite a different pattern, the splitlines being mainly transverse anteriorly
and gradually deviating to a diagonal
direction posteriorly. The next older specimen once more shows the anterior-posterior pattern. Another specimen at approximately the same stage of development as the two with deviant patterns
(Tappen, '57) shows an orientation that
is mainly antero-posterior, although some
of the splits assume a medial direction
near the midline.
It is very difficult to reconcile these variations with any theory which equates the
direction of orientation of split-lines with
growth direction. This is especially true
if the direction of growth of the bone is
considered to be mainly by self-differentiation independent of extrinsic factors, in
accordance with evidence pointing to this
conclusion in regard to the development of
gross structural features of bone (Murray,
'36; MOSS,'54).
There is reason to believe that direction
of bone growth at sutural margins of the
cranial vault may be primarily a response
to tensile forces resulting from growth of
the brain acting upon the intersutural
periosteum (Massler and Schour, '51). If
this is the case, the direction of growth of
the bone and the orientation of its structure may coincide here. The variability in
direction of split-lines in immature gorilla
cranial vault bones may in such a case be
caused by differences in the direction of
action of mechanical forces rather than
by genetically controlled intrinsic factors
of growth. (It may be significant that one
of the two deviant individuals was the
only specimen with a patent metopic suture.) It is likewise possible that the gen-
era1 confluence of split-line patterns with great majority of Haversian systems are
the axial direction of growth of long bones formed secondarily by concentric deposimay result from a similar identity with tion of layers of bone after resorption
the effects of mechanical forces, although spaces have been tunnelled out in previthe situation is probably more complex ously existing bone. It is possible that
non-Haversian vessels such as those traced
than in cranial vault bones.
2. Vascular orientation in bone as an in monkey femora are incorporated from
explanation of split-line orientation. The periosteum by developing bone. It is not
split-line experiment on monkey femora established that the periosteal vessels are
described in this article shows that the highly oriented, however, and the present
longitudinal orientation of the splits is in- study shows that these canals do not cordependent of the diagonal orientation of respond to split-lines.
the vascular canals. The proposition that
B. Criticisms of “mechanical
split-lines follow the course of blood vesforces” hypothesis
sels is clearly incorrect in this case.
A variety of arguments have been marThe experiment may be considered as
showing only a special instance in which shalled against interpretation of split-line
split-lines and vessels do not correspond, patterns as responses of bone to the action
since split-lines are usually made perpen- of mechanical forces. Moss (’54) does
dicular to natural surfaces of the compacta not offer any alternative explanation, but
rather than on cut surfaces, Split-lines on he makes one theoretical objection relethe periosteal surface of monkey femora vant to the present investigation. The gist
are also relevant to the problem. The of his argument is that the normal loadsplits are highly oriented axially on the ing on a bone “does not threaten the inperiosteal surface throughout the shaft of tegrity of the structure,” because “the orithe femur, as shown in figure 6, and can entation of the osteones does not become
be produced with either shallow or deep critical [italics by Moss] until the bone is
penetrations of the needle. These prepara- loaded beyond the normal range of imtions show no direct relationship micro- posed forces,” and therefore the orientascopically to the vascular canals, which tion of compact bone is difficult to explain
predictably penetrate diagonally to depths on the basis of “functional requirements”
beyond shallow split-lines. Microscopic de- (p. 378). The assumption is that orientatails of these preparations will be pre- tion of bone structure could not take place
sented in a subsequent article dealing with unless the strains were great enough to
effects of split-lines in relation to a variety create excessive deformation on the bones
of structural features of bone. It is quite involved, and therefore normal biomeclear from present evidence that surface chanical processes cannot be a cause of
preparations on the monkey femur like- oriented split-lines.
While the processes of bone formation
wise lend no support to the hypothesis that
split-lines are simply following the course and remodeling are far from being well
understood (Frost, ’63), there is no eviof blood vessels in the bone.
The idea seems to rest on a serious mis- dence to indicate that responses of bone to
conception of the relationship of vessels mechanical forces at the microscopic level
within a bone to its structural orientation. are necessarily related to actual strength
Evans and Goff (’57) believe “that prac- requirements helping to prevent fractures
tically all the Haversian vessels were orig- or excessive deformations of grossly obinally periosteal vessels which became servable structures. Available evidence
buried deeper and deeper in the cortex as tends to indicate the opposite, in fact. Rethe bone grew in diameter. Since the ves- sorption and bone formation apparently
sels of the Haversian canals run more or continue throughout life; new osteones
less longitudinally in the shaft of a bone show the same general orientation as those
the Haversian systems do likewise” (p. already in existence in adults, for exam73). The authors are paraphrasing Ham ple, as Moss (’54) himself states. Since
(’52). This is contrary to a large body of osteones usually form in areas where bone
evidence (Lacroix, ’51; Frost, ’63) that the existed previously, the resorption spaces.
that precede their formation must actually
weaken the bone temporarily. It therefore appears that Moss is attacking a
model of bone reorganization which has
not been advocated by anyone else, to my
knowledge. There is some evidence, however, that an additional volume of bone
may be built up in response to the action
of mechanical forces (Washburn, ’47; Enlow, ’62b). This would have the effect of
giving additional resistance to mechanical
Evans and Goff (’57) present several
lines of evidence which they believe render
untenable explanations of split-line patterns as responses to mechanical forces.
Their investigation centers on comparisons
of stresscoat and split-line patterns in the
same primate femora. The stresscoat technique uses a strain sensitive lacquer to
coat the bone. When the bone is subjected
to loading, a series of cracks form in the
lacquer transversely to the tensile strain.
Evans and Goff showed both tensile and
compressive strains in vertically loaded
primate femora by using an ingenious variation of the procedure. The same bones
were then decalcified and split-line preparations made.
Direct comparisons of the two kinds of
patterns from the stresscoat and split-line
techniques are not found in the descriptions of Evans and Goff, but they have supplied a number of photographs which are
quite revealing. The stresscoat cracks are
transverse to the axis in the shaft but on
opposite sides for tensile and compressive
strains. The patterns are at virtual right
angles to the major orientations of splitlines throughout the shaft. Evans and
Goff do not discuss this relationship. Instead they are interested in pointing out
that the split-lines do not correspond to
the theoretical trajectories they believe are
represented in part by the stresscoat
cracks. These “trajectories” are projected
(i.e., drawn in with ink beyond the actual
cracks) on a chimpanzee femur in which
stresscoat patterns were produced by loading vertically and by subsequently releasing the load. (Evans and Goff, ’57, figs. 18
and 19.) It is never explained why the
elaborately curved parts of the “trajectories” are not represented by any stresscoat
patterns, nor why the authors ignore their
own dictum : “Stress trajectories or lines
of stress’ cannot be seen in a bone and a
bone is not the type of body for which
trajectorial diagrams can be properly
One objection Evans and Goff make to
interpretations of split-line patterns in
terms of the effects of mechanical forces is
that the lines seldom, if ever, cross each
other. Consequently, they cannot represent theoretical stress trajectories, since
tensile and compressive trajectories cross
each other at a 90“ angle.
This is one of the reasons that I have
avoid using the term “trajectory” as a
synonym for split-lines in previous publications. However, it does not follow that
responses to mechanical forces cannot be
revealed by structural orientations, nor
that orientations brought about in this
manner necessarily duplicate trajectorial
diagrams. For example, the non-biological
stresscoat lacquer cracks only in response
to tensile strains. It is possible that the
structural basis of split-lines in bone also
represents a response to certain mechanical strains only, though the nature of the
response is probably much more complex.
Many of the clearest patterns of split-lines
that have been observed are in areas where
the forces acting on the bone must be
principally tensile in nature. In such areas
the split-lines are usually oriented in the
direction of tensile strains rather than at
the right angles shown by stresscoat
cracks. It is thus possible that both techniques are illustrating the effects of identical mechanical forces in the shaft of the
Evans and Goff also discuss briefly the
split-line condition in the human skull
vault, stating that the region is not subjected to any great mechanical stresses
and taking note of Benninghoffs (’25) observations that distinctly oriented splitline patterns are not found in this region.
They then state: “This seems strange if,
as some users believe, the split-lines indicate the strain orientation of osteones” (p.
70). The reasoning here is extremely obscure. The absence of mechanical stresses
accompanied by an absence of split-line
patterns does not seem to be conclusive
evidence that oriented split-line areas are
not responses to stresses where they do
Another difficulty in the interpretation
of split-lines in terms of mechanical
stresses cited by Evans and Goff is the poor
development of split-lines in areas of attachment of muscles and especially of
tendon attachments. For example, they
cite areas on the greater and lesser trochanters of the femur which show dots
rather than well-defined split-lines. This
accords with our own observations, particularly where areas of tendon attachment to bone are indicated. It should be
observed, however, that well defined splitline patterns lead up to these areas, which
are quite restricted in size. This suggests
that tendon attachments either bring about
special histological adaptations in the bone
or rearrange the bone architecture so that
it is oriented in the direction of pull of
the tendon at this localized area. The
problem requires further investigation at
the microscopic level. The study by Stilwell and Gray (’54) cited by Evans and
Goff is not relevant, since it deals with the
bony areas which have tendons sliding
over them, not areas of functional attachment of tendons to bone.
A final objection to the mechanical
forces hypothesis arises from experiments
by several investigators cited by Evans
and Goff, in which limb bones of dogs
were immobilized but no marked differences from normal limbs developed
in histological structure or distribution
and concentration of minerals. It therefore seems likely that immobilized limbs
would show longitudinal split-line patterns. This gives support to the concept of
Evans and Goff that there is a “force field
for growth” in long bones. The problem
requires further investigation. The biomechanical significance of a force field
for growth needs to be examined. It is
possible that it is essentially tensile in nature, as with the roofing bones of the
skull, and that there is consequently no
conflict with theories of mechanical forces
as an orienting mechanism in compact
1. A number of investigators have proposed that the oriented patterning of split-
lines in many areas of compact bone represents a response of bone tissue to the
action of mechanical forces. Two alternative hypotheses have been proposed. This
study offers evidence that these alternative
proposals are untenable as general explanations of split-line orientation.
2. The hypothesis that split-line orientations represent a response to the direction
of growth of bones is contradicted by observations of widely divergent patterns on
opposite surfaces of the ilum, scapula and
mandible of several species of mammals.
3. The hypothesis that split-lines represent conformity of bone tissue orientation to the pattern of blood vessels in
the bone is contradicted by experiments
showing no correspondence between splitline patterns and the orientation of microscopic vascular canals in the femora of
4. The evidence against the alternative
explanations tends to strengthen the “mechanical forces” hypothesis, in the absence
of other proposals.
5. Some of the arguments raised
against the “mechanical forces” hypothesis
have been examined and found to be unconvincing, while others merit further investigation.
6. There is still no conclusive demonstration of the validity of any proposals as
general explanations of split-line organization. The problem merits continued investigation.
Ahrens, H. J. 1936 Die Entwicklung der Spaltlinienarchitektur des knochernen menschlichen
Schadels. Morph. Jahrb., 77: 357-371.
Avis, V. 1959 The relation of the temporal
muscle to the form of the coronoid process.
Am. J. Phys. Anthrop., 17: 99-104.
BenninghofF, A. 1925 Spaltlinien am Knochen,
eine Methode zur Ermittlung der Architektur
platter Knochen. Verhandl. Anat. Gesellsch.,
34: 189-206.
Bruhnke, J. 1929 Ein Beitrag ZUT Struktur der
Knochencompakta an geanderte Beanspruchungen. Anat. Anzeiger, 63: 289-299.
Enlow, D. H. 1962a A study of the post-natal
growth and remodeling of bone. Am. J. Anat.,
110: 79-101.
1962b Functions of the Haversian system. Am. J. Anat., 110: 269-306.
Evans, F. G. 1957 Stress and Strain in Bones.
Charles C Thomas, Springfield, Ill.
Evans, F. G., and C. W. Goff 1957 A comparative study of the primate femur by means
of the stresscoat and the split-line techniques.
Am. J. Phys. Anthrop., 15: 59-77.
Frost, H. M. 1963 Bone Remodeling Dynamics.
Charles C Thomas, Springfield, Ill.
Ham, A. W. 1952 Some histophysiological
problems peculiar to calcified tissues. J. Bone
Joint Surg., 34A: 701-728.
Henckel, K. 0. 1931 Vergleichend-anatomische
Untersuchungen uber die Struktur der Knochencompakta nach der Spaltlinienmethode.
Morph. Jahrb., 66: 22-45.
Holmstrand, K. 1957 Biophysical investigations
of bone transplants. Plast. Reconst. Surg., 19:
Lacroix, P. 1951 The Organization of Bones.
Translated from the Amended French Edition
by Stewart Gilder. Blakiston, Philadelphia.
Massler, M., and I. Schour 1951 The growth
pattern of the cranial vault in the albino rat
as measured by vital staining with Alizarine
red “S.” Anat. Rec., 110: 83-101.
Mednick, L. W. 1955 The evolution of the
human ilium. Am. J. Phys. Anthrop., 13: 203216.
Moss, M. L. 1954 Demonstration of the intrinsic vascular pattern of compact bone. Am.
J. of Phys. Anthrop., 12: 373-380.
Murray, P. D. F. 1936 Bones. Cambridge University Press, Cambridge.
Pauwels, F. 1950 ifber die mechanische Bedeutung der grijberen Kortikalisstruktur beim normalen und pathologisch verbogenen Rohrenknochen. Anat. Nachricht., I : 53-67.
Seipel, C. M. 1948 Trajectories of the jaws.
Acta Odontol. Scand., 8: 81-191.
Stagg, F. B., and N. C. Tappen 1963 A device
for cutting parallel-ended blocks from decalcified long bones. Stain Technol., 38: 343-344.
Stilwell, D. L., Jr., and D. J. Gray 1954 The
microscopic structure of periosteum i n areas
of tendinous contact. Anat. Rec., 120: 663-677.
Tappen, N. C. 1953 A functional analysis of
the facial skeleton with split-line technique.
Am. J. Phys. Anthrop., 11: 503-532.
- 1954 A comparative functional analysis
of primate skulls by the split-line technique.
Hum. Biol., 26: 220-238.
1957 A comparison of split-line patterns in the skulls of a juvenile and a n adult
male gorilla. Am. J. Phys. Anthrop., 15: 4957.
1962 Block staining with hematoxylin,
gelatin embedding and serial sectioning of decalcified bone. Stain Tech., 37: 161-163.
1963 Split-line patterns in a series of
skulls of infant, juvenile and adult gorillas.
Am. J. Phys.Anthrop., 21: 135-151.
Washburn, S. L. 1947 The relation of the temporal muscle to the form of the skull. Anat.
Rec., 99: 239-248.
1 Split-line preparations on opposite surfaces of the same human
innominate bone, showing the strong deviation in major orientations
on ( l a ) external and ( l b ) internal surfaces of the wing of the ilium.
Split-line preparations on opposite surfaces of a dog scapula, showing
divergences of patterns of orientation i n some regions. In thinner
areas patterns on opposite surfaces can be seen through the bone.
2a, dorsal surface. 2b, costal surface.
N. C. Tappen
Split-line preparations on opposite surfaces of a human scapula,
showing divergences of patterns of orientation in some regions.
3a, costal surface. 3b, dorsal surface.
Split-line preparations on opposite surfaces of another human scapula. Patterns also diverge on opposite surfaces, but not identically
with those in figure 3. 4a, costal surface. 4b, dorsal surface.
N. C. Tappen
5a Split-line preparations on the cut surfaces of the cortical bone of the
shaft of a monkey femur which had been decalcified and bisected
5b Medial view of same monkey femur as in 5a, with the two parts
placed together. India ink from the split-lines shows through the
decalcified bone, revealing variable depth of penetration.
Split-line preparations on opposite surfaces of a fragment of the
shaft of a monkey femur.
N . C. Tappen
Microscopic serial cross-sections from mid-shaft region of a monkey
femur in which a split-line has been made perpendicular to the cut
surface, as in figure 5. The split conforms precisely to a cement
line, which continues beyond the depth of the split but outside the
field of view of the photomicrographs. The sequence is viewed from
distal to proximal. The trend of all the vascular canals is inward.
Some canals can be observed passing through cement lines, which
remain constant in relation to periosteal and endosteal surfaces for
much of the length of the shaft. Hematoxylin block stain. x 30.
N. C. Tappen
Higher power view showing detail of passage of a vascular canal
through a split-line made perpendicular to the cut surface of a monkey femur. Split-line conforms to a cement line. The canal is
penetrating from the right, or periosteal, surface; the sequence is
viewed from distal to proximal. Hematoxylin block stain. X 89.
N. C. Tappen
44 1
Без категории
Размер файла
1 430 Кб
orientation, examination, explanation, compact, split, alternative, line, bones
Пожаловаться на содержимое документа