STUDIES ON THE FEMUR VI. THE DISTAL PART OF THE DIAPHYSIS N. W. I N G A L L S AND M. €I. GROSSRERG Anatomical LabolatoI-j, Tl'estern Reserve University, CkUeland. O7iio FOUR FIGURES There is preserved in the Hamanii Alusenni a most remarkable specimen, represented by the pelvis and the long bones of the lower extremities. The most striking and unusual features a r e those found i n the femora, where the distal portion of the shafts a r e much elongated and at the same time enormously widened. These bones, but more especially the femora, were described very briefly by Doctor Hamann in 1910. A more detailed account of these unnsual bones mill be found elsewhere (Ingalls, '32). It was, indeed, the extraordinary conditions presented by these bones which directed our attention to the lower end of the femoral diaphysis. I n a n earlier paper ( '24), based on a study of 200 femora, from 100 male whites, it was pointed out that the diaphysis of the femur is distinctly morc. variable than the epiphyses. This was taken to indicate that the various adjustments and other responses l o different influences, are carried out by or mainly in evidence in, the diaphysis rather than in the epiphyses. The latter, as integral parts of some joint, may adapt themselves to a considerable extent by joint movements, but the diaphysis plays a quite different r6le, a n d its more important functions would appear to be better served by greater plasticity and ease of modification, i.e., by a greater variability. Not only is the diaphysis more variable than the epiphyses hut the transverse measurements of the shaft a re more variahle than 475 AMERIOAN J O L H N A L OF PHYSICAL AKTHROPOLOGY, VOL. S V I , NO. 4 A P B I k J U X E , 1932 476 Ii. W. I S G A L L S A S D M. H. GEOSSBERG those taken in tlie sagittal plane, i.e., at right angles, and it is with this great variability of the transverse diameters of the distal p a r t of the femoral diaphysis that we are particularly concerned a t present. I n order to gain more detailed and accurate information on what seemed to be the more significant characters of the lower end of the shaft, a number of special measurements were taken, as will appear below. The material used in this study is identical with that which formed the basis of onr first paper. W e shall also make use of certain of the measurements noted there, retaining at the same time the designations and numbers previously employed. Details of technique not included here will be found in our earlier paper to which reference has already been made. The femoral dimensions which will be particularly considered a t this time a r e as follows. N o . 6 . Diaphyseal length. Measured in the axis of the shaft, f r o m the summit of the trochlcar pass below to the intersection of the axes of neck and shaft above. 12. Least sagittal diameter, i n f e r i o r diaphysis. This measurement, taken from Martin ('14), is chiefly of importance since it forms the starting point from which nos. 13-19 a r c determined. I t s location along the axis of the bone fluctuates somewhat, but on the average it is about 2 cm. or even less, above the upper limit of the inferior hrticular surface in the mid-linc. It is not a very satisfactory measurement to take, but its disadvantages do not seem to be serious, at least for our present purpose. 13. T r a n s v e r s e diameter at t h e same level, traizsverse, I . At the level of 12 and at right angles to the axis of the shaft. This measurement is somewhat similar to Pearson's popliteal width, but is taken at a much lower level; it is more like the determination used by Klaatsch ('01). No. 13, or I, is a diaphyseal dimension, while nos. 26 and 27 (v.;.), a r e concerned only with the lower epiphysis (figs. 1 and 2). No. 13 is of especial interest because it is the first, and lowest, of a series of special transverse measurements (v-i.) ; in the STUDIES O K THE FEMUR. VI 477 femur a t birth it lies at the junction of diaphysis and epiph y sis. 14 to 19, inclusive, referred to hereafter as 11-VII, a r e special traitsverse measurements at right angles to the axis of the shaft and at equal intervals above 13, o r I. I n order to keep this interval always in terms of the size, length of the individual bone, so obviating as f a r as possible discrepancies due to long and short femora, the distance between any two successive measurements, I to VII, was taken a s one fifteenth, 1/15, of the diaphyseal length, no. 6. I n our series of 100 the mean diaphyseal length, right and left, is 391 mm. and the mean interval is therefore a trifle over 26 mm. This rather small interval provides a n accurate check on the changing form of the bone, and by using seven of these levels the distal half of the femur is covered; indeed the last and highest level, VII, i s practically identical with the pilasteric level, for which it has been substituted in the fetal series. 20, 22, and 23 are sagittal measurements, made at the levels of 11,111,and V, from the most prominent point on the ventral surface of the shaft to the mid-line behind; they correspond with Manouvrier 's line m p in the popliteal region. 26. Epicondylar breadth, this and 27, i n f e r i o r articular breadth, a r e epiphyseal characters ; they a r e measured parallel with the infracondylar plane, but, due to the ordinary obliquity of the femoral shaft, they a r e not parallel with I to VII. Although there are present in the lower end of the femur, some of its most characteristically human features, this part of the bone, except for the condylar region, knee-joint, has not received the attention which it deserves. As regards the distal portion of the diaphysis, the popliteal region, it has long been known that it is subject to considerable individual, if not racial variation, particularly in its transverse diameters. Actual measurements of this part of the bone, appear to be limited, however, to the popliteal sagittal and transverse dimensions, and their related indices, and, to a much less extent, to the determination of the popliteal length. 478 N. W. INGALLS A N D M. H. GROSSBERG The descriptions given by Dubois of the Trinil femur in which he stated that “the popliteal space is less developed, convex in its middle, so that at this height the shaft is almost round instead of flattened” (quoted from Hepburn, ’96) were followed by extensive investigations of this region, Manouvrier ( ’95), Hepburn ( ’96-97) among others, in which the popliteal index played the most important part. Indeed, most of this work was concerned with the form of the femoral shaft at the level measured, whether the posterior surface, planum popliteum, was flat or convex or concave; since Dubois had maintained that a convex popliteal surface was peculiar to Pithecanthropus, and that it did not occur in human femora. At present, however, we are especially concerned, not with the convexity of the popliteal surface, in regard to which Dubois was in error, but with the width of the bone, naturally in respect to its antero-posterior dimensions, their relative values being expressed by the popliteal index. The popliteal index is represented by the sagittal diameter, x 100/transverse, both measurements being taken at the same level. What this level should be is a matter of considerable importance, on account of the great variability in the form of this part of the bone, more particularly its transverse dimensions. I n the case of the pilasteric and platpmeric diameters the conditions are much simpler. In the present series of 100 femora we followed the directions given in the first edition of Martin’s Lehrbuch (’14) which calls for the least sagittal diameter of the inferior of the diaphysis, measured in the mid-line of the bone and about 1 em. above the articular surface of the condyles. This is Xartin’s measurement no. 11 (our no. 12). I n the second edition of the work (’28) the directions stand unchanged except that the distance above the condyles is now 4 em. instead of 1 em., but perhaps this 1 em. should have been 4 em. As noted above this sagittal determination is not entirely satisfactory, but it has the advantage in our present series, that it is at a distinctly lower level than that at which the popliteal dimen- STUDIES O N THE FEMUR. 479 VI sions a r e coiiimonly taken. On our own cases this level is not over 3 em. above the articular surface i n the mid-line, and less than this above the surface on the lateral condyle. Manouvrier, and others following him, Hepburn and Pearson have taken their popliteal level 4 em. above the articular surface on the lateral condyle, and Rlanouvrier was very wisely inclined to vary this distance for long and short bones. This 4 cin. level corresponds very closely with the level of our special transverse 11. It is evident, we think that if popliteal TABLE 1 Data 01) 100 p a w nude while fernom. Right and lefi combined. ezcept the range, are n mean of right and !cft means .. All valurs, - _- BAIGE NO. Max. _. G 459. 12 13 14 15 16 -- I I . 17 pecial transverse, PI ISpeeial transverse, T T ISagittal, level of I1 (Sagittal, level of 111 isagittal, level of V 26 ,Epicondpkr breadth art. surface, 27 1 breadth 18 19 20 29 23 __ I' Nln. 1341. 41.5 I 24. 28.78C 28.35, 31.615 30.354 28.87-C 83.462 ' .14, 2.072 .1@ 7.21lrt.34 .I4 2.042 .I0 ' 7.202.34 .19' 20: .l7 .2G I I 2.77-fr .13 i 8 . 4 8 2 2 6 , 3.912 .19 __ '1 8.772.42 .13 8.i22.42 2 . 3 2 2 .12 I 8.74&.42 3.912 .19 I 4.692.32 2.64rt 55. 48. 43.5 38. 34.3 34.5 40. 41. 40. 93. I 1 _4.952.24 _.__ 92.5 1 42. 33. 28.5 25.5 24. 23. 22.5 26. 24. 1 23. i 70. i 68. measurements a t only one level are to be fairly comparable, and therefore of value, the level selected should be in terms of the length of the shaft, diaphyseal length, rather than any constant or arbitrary distance above the condylar surface. These discrepancies and technical shortcomings, however, need not concern us particularly since for our own material there a r e a sufficient vzurnber of levels and at short e m u g h iqttewals to bring out the features in which we a r e interested. In table 1, there has been brought together the data from 200 male white femora, a s f a r as it relates to the present ques- 480 N. W. INGALLS AND M. H. GROSSBERG tion. In figure 1, the mean values of certain measurements a r e indicated graphically, both for the adult and fetal bone, together with the corresponding coefficients of variation. The left hand outline in figure 1, represents the adult femur, reconstructed from mean values, the numerals within Fig. 1 Outlilies of the male white femur in the adult and a t birth. Based on mean measurements of 100 pairs of adult, and 10 pairs of fetal bones, both bones in terms of the same oblique length. The numerals within the bones indicate the number, level and direction of the measurement. S t the left of each outline, there are indicated the coefficients G€ variability opposite their corresponding measurements. The vertical lines represent the mean Variability of the seventeen femoral characters listed in table 1 of “ T h e femur a t birth.” indicate the transverse measurements, while to the left there have been plotted the corresponding coefficients of variation. The two lowest measurements a r e made on the epiphysis and a r e characterized by their relatively low variability. No. 27, the articular breadth, has a coefficient of 4.98, while the epicondylar breadth, no. 26, immediately above it, is, if anything, even more constant, 4.69. Indeed the epicondylar breadth STUDIES O K T H E FE,MUR. VI 481 is apparently the most constant, least variable dimension which can be taken on the femur. This is of special interest and significance, because measurements taken only a few centimeters higher up on the bone, but in the same plane, show a very high degree of variability, and one which is surpassed only by some of the angular measurements. It is in this region, in the most distal part of the femur, that the difference in variability of the diaphysis and epiphysis is most accentuated. Measurements I is at right angles to the axis of the shaft and is the lowest of the special transverse measurements on the diaphysis. The mean interval between the succeeding levels, I1 to VII, is 26 mm. I n figure 1 it i s evident that between 26, the epicondylar breadth and I, the lowest transverse diameter of the diaphysis, there is a very sudden and striking increase in the variability. I n the former the coefficient of variability is 4.69 while in the later it is 10.20, more than tmice a s great. These figures represent the mean of the values for the right and left sides, and if the coefficients for the left only had been used, the difference between epiphysis and diaphysis would have been even more pronounced. As a transitional region between the massive joint ends and the relatively slender shaft, one might have expected that the lower end of the shaft would show considerable fluctuation i n its dimensions, this variability however, i s not confined to the immediate neighborhood of the junction of the shaft and epiphysis, but is continued upward, essentially unchanged for a considerable distance. I n this respect it is unlike the fetal bone. From level I to V I inclusive over an extent of nearly 80 mm., the variability is practically constant, while above this it gradually diminishes toward the pilasteric region. The mean value of the coefficients from I-IV inelusive is 10.14; at V it has fallen to 8.38, while the mean for V I and V I I is 7.21, not materially different from the pilasteric transverse which has a variability of 7.27. From the foregoing we may conclude that the most distal part of the femoral shaft over a mean extent of 100 mm., 482 T S . W. IXGALLS A N D M. H. GROSSBERG from below level I to above level IV, is characterized by a high and constant variability in its transverse dimensions. This 100 mm., on the average, is a little more than one fourth of the diaphyseal length a s we have measured it, but since this length includes a part of the lower epiphysis, the extent of diaphysis involved would be definitely more than one quarter of its real length, however measured. I n contrast with these transverse measurements, we find that the sagittal dimensions taken at the same levels are much more constant. Of these, however, I, or no. 12 is somewhat of a n exception, due in part, no doubt, to the technical uncertainties involved in its determination, its mean variability is 9.54. The sagittal diameters at levels 11, III and V, the only ones we have determined, are almost identical, the mean coefficient being 8.74. At higher levels in the shaft the sagittal dimensions are even more constant, the mean pilasteric having a coefficient of '7.69, the platymeric 6.98. Only in the middle of the shaft, in the pilasteric region and just below it, is the sagittal variability greater than the transverse, while the platymeric transverse is less variable than comparable measurements at the lower end. I n figure 2, the form of the lower half of the femur is represented schematically, but without regard to the obliquity of the shaft and certain other minor asymmetries in this region. In the four outlines shown here the diaphyseal length has been taken a s equal; i.e., 100 per cent in all cases. I t is apparent that the fetal femur is characterized by a very broad, massive condylar region, in proportion to the length of the bone, and by the marked flaring of the lower part of the shaft. Nos. 308 and 600 are special cases taken from the series of 100 pairs, the former representing one of the narrower, pinched-in types of femur, the latter being of the broad flaring kind. It will be noticed that these two special cases and the mean, between them, are most alike in the epiphyseal region, 26 and 27, and in the middle of the shaft. No. 600 has a relatively narrow epiphysis, in spite of its breadth above, since it is an unusually long bone, and the STCDIES O N THE F E M U R . TI 483 proportions in figure 2 a r e on the basis of the diaphyseal length. Thc differences represented here a r e obviously most marked in tlie lower p a r t of the shaft, in h t p a r t which is distinguished from the others liy its greater variability. There a r e indications also that in the broad types of boiie tlie tleviatioii from average conditions is most marked low Fig. 2 Scliematic outlines of lower end of femur, represeiiti~?gmean adult a n d mean fetal conditions, and also eases nos. 306 and 500; all on the basis of the same diaphyscal length. The usual obliquity of the shaft has been disregarded, therefore 26 and 27 appear parallel t o the higher planes of mcasuremciit. I n order to keep the contours separate, 110. 6@0 lias not becii carried upward a s f a r as the others. clowii, while in tlie narrower, more cylintlrical bo~ies, this difference is more coiispicuous a t a somewhat higher level. Examples of some of the different types of femoral shaft a r e shown in figure 3. All of these bones, escept E 386 and the two Keanderthal femora 011 the right, a r e taken from our series of 100. The central specimen, 110. 369 although shorter than the others, represents very closely average conditioiis, as f a r as the lower elid of the shaft is concerned. A M E R I C A N T C U R N A L O F I ’ I I Y S I C A L A N T H R O P O L O G Y , VOL. S V I , N O . 4 Fig. 3 Gcrios of' 1)31ies to slio~v the v:iryiiig configurntion of tlic l o ~ r r rpnrt of the fcinornl slinft. No. 369 ni:ig bc t:tkeii as rrprwentii!g t h e i i i e a i i f o r oiir series, R S f:ir ns the proportions of tlic l o w ~ iend ~ of t.lic diiipliysie :ire conccwied. I.: 386 oii the left. is tlie niiomnlous f r l n u r drscribcd at. t.lic p i i d of this :irticlc. S is :t cast of the Spy, Nc:i!idcrtli:~l f c ~ n i i r , I> that oF 1)iisscldorf. STUDIES ON THE FEMUR. VI 485 I t may be noted here that it was only in regards to a very few especially wide femora, particularly no. 600, that the question arose of excluding them from the series. The exclusion of any bone, however, whether wide or narrow, would have involved the setting up of some staiidards on the very points on which we were engaged in procuring information. Although no selection whatever has entered into this series, as far as it relates to the lower end of the bone, nevertheless we feel that in one or more cases, most probably in no. 600, we a r e dealing with bones which a r e not altogether normal, but a r e transitional or border line cases. As noted earlier, the popliteal index has been mainly employed in its relation to the convexity or concavity of the popliteal surface. O u r present interest in this index is rather as an expression of the width of this part of the femur i n terms of its antero-posterior dimension. If the index is 100 both popliteal diameters a r e the same, and the lower the index the broader a n d more flattened is the bone, the popliteal platymeria of Manouvrier. Computed on the basis of our lowest popliteal dimensions, at the level of I, the index of means in our series, combiniiig right and left, is 54; but this level is below that most commonly used, and therefore we shall compare the index at o u r level I1 with the index as it is usually given. A t this point the index i n our material is 76.4; Pearsons finds an index of approximately 71, while Hepburn’s modern British run from 66 to nearly 90, with an average value of 78.1. Similar indices computed €or levels 111and V shorn a gradual rise in the index, 84. and 96. until it becomes 103.5 as the pilasteric index. The popliteal indices of no. 308 and no. 600, taken at levels 11, I T and V, show very well the relation between the index and the width of the bone. I n 305 the indices are, in round numbers, 80, 91, and 100; for 600 they a r e 71, 74 and 83. Pearson’s popliteal-bicondylar, o r sofit index, popliteal width Y 100/bicondylar width, is especially useful in this connection, in this index a higher figure means a greater 486 N. W. INGALLS A N D M. H. GROSSBERG popliteal width, a broader femoral shaft above the condyles. I n the present series the index is 49.5, computed at level TI, a n d Pearson's results a r c almost identical, 50. In no. 369 (fig. 3 ) , which represents about average conditions, this soffit index is also 49.5; in no. 308 it is only 41. but in no. 600 it is 61. We have also computed what may be termed a second a n d third soffit index, at the levels of I11 and ITT, and an additional index using the pilasteric instead of the popliteal width. F o r these three indices the series of 100 show the following values; 43, 39, and 34; f o r no. 369 they a r e 43, 38 and 33; for no. 308, 35, 21 and 32; and for no. 600, 53, 48 and 32. H e r e again we find a remarkably constant relation between the middle of the shaft and the condylar region, a relation which is essentially the same in the Neanderthal and Sp y femora and also i n E 386. Among his numerous femoral measurements, Pearson determined what he called the popliteal length. We have no d a t a on popliteal length f o r our own material, we mention i t here because we have this measurement on E 386, but more particularly because it may indicate that the part of the femoral shaft with which we a r e concerned, is not only unusually variable in its transverse diameter, but also in its length, ik.,as measured in the axis of the shaft. 1F-e shall return to this in discussing the conditions presented by E 386. On the right in figure 1. certain characters of the femur at birth have been represented f o r comparison with adult conditions. Not only is the mean variability of the fetal femur considerably lower than that of the bone during later life, 4.69 a s compared with 6.44 f o r the same measurements, but the relative variability of some of these characters is not the same in the adult a s it was at birth. I n the fetal series, the epicondylar breadth, 26, and the inferior articular breadth, 27, a r e closely similar in their variability, 5.02 and 5.14, values which are only a little in excess of those shown by adult b0nes.l I n the fetus as in 'In our paper on the femur at birth the coefficient of variation for the epicondylar breadth, no. 12, table 1, was given as 3.021. This should read 5.025. In the text, a few lines below, the mean variability of 17 measurements was noted as 4.879. This should be 4.961. STUDIES ON THE FEMUR. VI 487 the adult there is a n increase in variability in passing from 26 to 1, but the rise is not so conspicuous. The coefficient for I at birth is only 6.78 while later in life it becomes 10.20. Above the level of I, however, there is a steady decrease in the variability, quite in contrast with adult conditions, where this decrease began much higher up, at the level of IV. Comparing the fetal and adult conditions, we find that the variability of the epiphysis, 26 and 27 does not change materially during the growth a n d development of the bone, but that for the distal p a r t of the shaft, in its transverse dimensions, there is a very striking increase in the variability. The mean variability in dimensions I-VII inclusive is 9.05 in thc adult and only 4.40 in the fetus; for the four lower levels, I-IV inclusive it is 10.14 in the adult, but oiily 5.05 at birth. That I is the most variable of any of these characters in the fetus, and that, in a way, it stands alone, is due, we imagine, to the fact that it lies so close to the epiphyseal line, at the junction of the epiphysis and diaphysis. Why V in the fetus should appear so much more variable than either IV or VI. we cannot s a y ; we have checked over our original measurements and can find nothing to account for it. The higher coefficient here may be the result of some error o r be due, in part, to the shortness of the series. A comparison of fetal and adult conditions in the lower end of the femur reveals a remarkably stable and constant epiphysis, joint, and a n equally remarkable variability, or even instability, in the immediately adjacent diaphysis, or shaft. Although a t birth the lower part of the shaft except, at level I, is no more variable than other femoral characters, there is, we believe, a very definite latent variability in this p a r t of the bone, a variability which gives evidence of its presence and degree only later, during growth and differentiation, and in response to a great variety of influences or stimuli, which may or may not confine themselves within normal and proper limits. The final form of this part of the femur is brought about by extrinsic factors acting upon a certain hereditary makeup which is itself unusnally plastic and variable. 488 B. R. IXGALLS AND M. H. GROSSEEILG I n regard to the various popliteal indices at birth little need be said. The popliteal region is relatively broad at this time but this peculiarity seems to be in a lardoe measure merely secondary to the great width of the condylar region, figure 2, rather than representing any precocious or extreme development of essentially human features. The relatively high degree of variability evidenced by the distal third of the femoral shaft is to be looked upon as the direct result of the comparatively recent and characteristic modifications which have come about in this part of the bone. These changes are for the most part, if not exclusively, related to the assumption of a really upright posture, and include alterations in the musculature and joint mechanisms as well as in the shaft of the femur. I n the process of this gradual remodelling of the femur, to meet new static and kinetic conditions, it is the lower end of the bone which has been altered most extensively, while the upper end has departed less from its primitive condition. The shaft, as distinguished from the joint ends, has become much elongated and further strengthened and improved by the widening of its distal portion and the pilastering in the middle. That the most important and conspicuous changes have taken place in the lower femur, including the knee joint, would indicate that the assumption and maintenance of a functionally adequate erect position, has been largely a question of evolving a knee joint fitted for and able to carry out its newer functions. Especially important in this respect was complete extension of the knee together with adequate assurance of its stability and usefulness in this position, without impairing in any way its mobility or general utility in various degrees of flexion. To this end the whole condylar region has been markedly increased in size, particularly from before backwards, and the external condyle has become especially prominent. I n the shaft just above, the transverse mass of the bone has been increased and the lateral border of this region has been thickened in addition, where it transmits the superincumbent weight and strain from the shaft and pilaster to the body of the lateral condyle below. STUDIES OX’ T H E FEMUR. \‘I 489 That these alterations have left the epipliysis, or joint, in a fairly stable and constant condition, while the diaphysis above has remained o r even become more plastic and variable, may be explained by the natural mobility of the former, as a joint surface, where a great variety of adjustments a i d adaptations can be cared for by the simple expedient of joint movements. The relatively low variability of the joint end may be further guaranteed, as it were, by the necessity of keeping the character of the joint surfaces and other important functional features within comparatively liarrow limits. I n the shaft, on the other hand, and particularly in its lower end, which mediates between the joint and the rest of the bone, adjustments or modifications of any kind must, manifest themselves in the form and structure of the part, rather than in a n y form of movement. The lower femoral shaft is variable, partly on account of the relative newness of certain of its characters, but also because of its plasticity and its ability to adjust and correlate the rest of the body with the knee joint, due to its intimate proximity and association with that joint. TJnlilre the joint proper, the characters of the lower shaft are determined largely by the conditions, iiormal o r otherwise, obtaining in any particular case, they are, to a considerable estent, acquired rather than hereditary, the result of the operation of external more than of internal factors. In the gradual development of the human lower limb, with its many characteristic features, it is obvious that the soft parts, and particularly the musculature, has been involved a s well as the skeleton. To what extent or in just what way, the various muscles, or muscle groups, like the quadriceps, a r e correlated with different types of femur, either in the pilasteric or in the popliteal regions, is a s yet by no means clear, and much of the evidence is indefinite or even contradictory. Unfortunately the investigation of the soft parts is not a s easy or pleasant a s the examination of dried bones, and from a statistical stand point the difliculties a r e enormously enhanced. The complete story of bone, either as ;-1 490 N. W. I N G A L L S A X D M. H. GROSSBERG tissue o r as a n organ, will never be written uiitil it is studied in its natural environment, as a living, and extremely scnsitive p a r t of the animal economy, primarily adapted to withstanding certain physical strains, but a t the same time nnusually responsive to a variety of other influences, chemical, metabolic, etc. The conspicuous variability of the lower p a r t of the femoral shaft, which we look upon as conditioned by its relative newness and by its peculiar and important relations to the knee joint, and through this to the erect position, has another aspect in which this variability appears more as a definite instability o r lability o r even a positive Vulnerability. - h i this respect one is reminded of another characteristically human a n d recently modified portion of the skeleton, also of significance f or the upright posture, the lumbar spine, where morphological variations or frank defects are very common, and where functional weaknesses o r shortcomings give evidence of some inherent instability o r structural inadequacy (Willis, ’24; Ingalls, ’31). As backache and other related disabilities and infirmities, a r e one of the numerous byproducts of the erect posture, so, likewise, is the natural and of ten unconquerable proneness of the knee joint to assume a flexed position, the proverbial “weak kneedness.” The peculiar vulnerability of the lower femoral shaft, its susceptibility to various influences, is shown by certain border line cases between normal variation and frank pathology, by the localization here of certain abnormal conditions, as well as by the involvement of this region in a variety of disturbances which have little o r nothing to do with the normal functions of the femur. W e shall refer to this aspect of femoral variability more in detail elsewhere. It may be noted i n passing that Ilanouvrier was somewhat in doubt as to the normality of this part of the Trinil femur, his suspicions being prompted by the existence of gross pathological lesions, of quite another kind, in the upper part of the shaft. STUDIES 011'THE FEMUR. VI 49 1 As illustrative of the relation of variability to pathology, of' the overlap, as it were, between the normal and the abnormal, we shall give a brief account of some of the more significant features displayed by specimen Ii: 386, mentioned a t the beginning of this article. I t is an especially instructive case because the usual anthropological determinations, without illustrations, would have betrayed little of the real character of this bone (figs. 3 and 4). Fig. 4 Schematic representation of the lower elid of E 386 and of the mean conditions in our series of 100. On the left in terms of measuremeiit 110. 26, the epicondylar breadth; 011 the right in terms of no. F, the diaphyseal length. It i s evident that the condition present in E 356 i s limited t o the lower part of the shaft. For the right femur of E 386, a few of the more important measurements may be listed f o r comparison. The maximum and obliques lengths are 514.5 and 511 mm., respectively, the trochanteric and oblique trochanteric lengths a r e in proportion. Although this is a large and long bone all of these dimensions a r e exceeded in our series of 100. I t s weight is 680 gm., but some bone substance has been 492 N. W. I X G A L L S A N D M. H. GROSSBERG lost, and the original weight would be very close to if it is not actually greater than the maximum €or the 100 series, 695 gm. The bone is unusually heavy, rather than long. The platymeric a nd pilasteric diameters a r e large, but well within our range; the platymeric index is 88.6 as compared with our mean (100 cases) of 83.5, the pilastcric indices are essentially the same. T h e vertical and horizontal diameters of the head 51 and 50 mm., a r e only a little above the average, although the ellipticity is greater than usual the index being 98.0 instead of 99.6. The condylar region is large but in nowise exceeds our pervious limits, the lateral condyle shows a greater length, than usual, when cornpared with the medial condyle. Beyond what has just been noted, the indices indicate nothing beyond a rather long bone, with a relatively small head and a narrower platymeric and broader pilasteric regions when compared with average conditions. I t is when one investigates the lower p a r t of the shaft ot' E. 386, that its peculiarities become evident. Close to the epiphysis the shaft is still not unusually broad as indicated by our popliteal index a t level I. The index in E 386, is 51.4, the mean f o r 100 is 54, but in no. 308 the index is only 50.5. A t levels TI, I11 and V liowever the story is very different, in the 100 series the corresponding indices are, in round numbers, 76, 84 and 96, while in E. 38G these a r e much lower, 60, 58 and 68, or in other words the bone is much broader in proportion to its sagittal diameter, but at the pilasteric level the indices a r e practically identical. A t Manouvrier's level, 4 em. above the lateral condyle, the index in E. 386 is about 60, while Pearson's femora showed an index of nearly 71. The three soffit indices noted above show very clearly the great width of the shaft when compared with the bicondylar width. In the series of 100 these three indices, first, second and third we called them, a r e 49, 43 and 39, in E. 386 they a r e 69, 67, and 63. I n these indices more than anywhere else there a r e revealed the real peculiarities of this abnormal bone. I f , however, one computes a n index by substituting the pilasteric transverse f o r the STUDIES O N THE FEMUR. VI 493 popliteal width in the soffit index, then there is nothing to distinguish E. 386 from the average femur, in both cases the index is the same, 34. Mention was made above of Pearson’s popliteal length, (l.c., p. 9 ) taken from the point of bifurcation of the lips of the linea aspera to the upper limit of the intercondylar fossa in the midline. Although the upper end or apex of the popliteal surface may be somewhat uncertain in its exact location, there is no doubt that very unusual conditions obtain in E. 386. Pearson finds that the mean popliteal length is 121 mm., but in our case it is from 190 to well over 200 mm. depending on the determination of the apex. TJsing this length Pearson has computed a number of indices and a comparison of his results with ours indicates a most extraordinary lengthening of the popliteal space in our case. Briefly put and in round numbers, in his material the popliteal length is 32 per cent of the diaphyseal, in ours it is 45 ; the bicondylar breadth is, in his cases 66 per cent of the popliteal length, but only 44 per cent in E. 386. As f a r a s it goes the foregoing suggests that in E 386 more than the usual contribution to femoral length has been made at the lower end, that not only is this bone exceptionally wide distally but that it is also unusually long, PO to speak, in its distal part. A frontal section through the entire length of the left femur reveals an excessive amount of cancellous tissue, the real marrow cavity is very much shorter than usual, being confined to the upper p a r t of the shaft. I n addition to what has been said it should be noted that the conditions described here a r e bilateral, both femora being involved in identically the same may and to the same extent. Important also is the fact that both tibiae a r e long and their upper ends are markedly expanded, much like the lower ends of the femora. Apparently the disturbance, in this particular case has affected the long bones adjacent to the knee, but the effects are much more in evidence in the femora. AMERIOAN JOURNAL OF PHYSICAL A N T A R O P O L O G Y , VOL. XVI, NO. 4 494 N. M’. I S G A L L S AND M. H. GROSSBERG It is altogether likely, we believe, that an investigation of the upper end of the tibia mould reveal conditions of variability and sensitivity much like those in the femur, and for substantially the same reasons. I n this case the tibia is very long, possibly somewhat out of proportion to the femur, as indicated by the high tibio-femoral index, which is nearly 87. Here, as in the femur, there may be evidence of iucreased length in the neighborhood of the knee, rather than a more even distribution of growth throughout the bone. I n none of these bones, either femur, tibia or fibula, is there any indication of inflammatory processes o r any tissue pathology, the bones a r e clean, hard and smooth. The most significant feature in this abnormal femur is the fact the disturbance, whatever it may be, is sharply localized in the lower p a r t of the shaft and that the transverse rather than the sagittal dimensions have been altered. This is exactly the region of greatest variability, and apparently of greatest sensitivity or vulnerability also, and it is this p a r t of the bone therefore, that has been especially picked out by some as yet unknown influence (probably metabolic or endocrine). As far as we know this is a unique specimen; i n our series of 100 there a r e a few cases in which the lower shaft is unusually broad, e.g., no. 600, but nothing even approaching the conditions seen in E. 386. What map very well be a comparable o r related condition was found by Wood Jones i n ancient Nubian material. “ I n body 45: 240, an adult man of the archaic period from the large cemetery a t Delimit, the left femur and tibia had undergone a very remarkable change. The lower half of the femur and the upper half of the tibia have become greatly enlarged and thickened” (Elliot Smith a n d Wood Jones, ’10, p. 289 and pl. XLVIII, fig. 4). In this case the surface of the bones was roughened and irregular, but the knee joint was not involved in the same process. Apparently the condition is unilateral here, altliougli we caiinot be certain, and there is the tentative suggestion of its luetic origin. STUDIES ON T H E FEMUR. VI 495 L I T E R A T U R E CITED HAMAPTN, C. A. 1910 An abnormality in the form of the femur. Clevelalid Medical J., vol. 9, pp. 710-i12. HEPBURN,DAVID 1696 The Triiiil femur (Pithecaiithrolius erectus), contrasted with the femora of various savage and civilizeil races. J. Ailat. and Pliys., V O ~ . 31, PI>. 1-17. 189i The platynieric, pilasteric and popliteal collection of femora in the Anatomical Museum of the University of Edinburgh. Ibid., VOI. 31, pp. 116-156. INGALLS,pi. W. 1924 Studies 011 the femur. I. General characters of the femur in the male white. Am. J . Phys. Anthrop., vol. 7, pp. 207-23.7. ____ 1927 Studies on tlie femur. V. The femur at birth. Am. J. l’hys. Anthrop., V O ~ .11, pp. 107--121. ~1931 Observations on bone weights. I. .4m. J . .411at., vol. 4S, 1)11. 45-98. ____1932 Studies ou the femur. VII. KLAATSCII,H. 1901 Die wichtigsteii Variationen a m Skelet der freieiii untereii Extremitat des Menschen und ihre Bcdeutuiig f u r das S b s t a m niungsproblem. Ergebnisse d. Aiiat. u. Entvick., Eld. 10, S. 599-719. MANOUVRIEK, L. 1895 Deunirine etude sur le “Pitheeantliropiis t’rectus” comme precurseur presumk de l’liomme. Bul. %lela Soc. d’Antlirop. de Paris, series IV, T. 6 . MARTIS, H ~ ~ D O L1914 F Lehrbuch der hnthropologie. .Jeiia, 2te Aiiflage, 1928. P E A R S O X , KARL, A N D JULIA BELL 1919 A study of the long bones of tlie Euglish skeleton. P a r t I. The femur. Drapers’ Co. Research M e w , Biometric series X, London. SMITH, G. >;LI,IOT, A N D F. WOOD JONES 1910 The archeological survey of Il-ubia, Report for 1907-08. 1‘01. 11. Report on the humau remaii,s. Cairo. WILI~IS, T. A. 1924 Backache froin vertebral anomaly. Surgery, Gynecology and Ollstetrics, Map, pp. 658-565.