A. B. Emmons 
45 
The fiat pelves, 56 in all, show that at least slight contraction of the inlet 
is to be expected in one pelvis in every four, while moderate to serious narrowing 
of this all-important space is to be expected about once in twenty pelves. 
3. The outlet showed wide variation in size and shape. By the use of the 
" fetal head " measure it was determined that no pelvis, in all probability, was 
too small to pass the average fetal head. It was found that a narrowing of the 
inter-tuberosities diameter was the most important single factor in reducing the 
size of the available space. Next in importance was the shortening of the space 
behind this line, or in other words a reduction of the posterior-sagittal diameter, 
found with but slightly greater frequency in pelves in which the sacrum contained 
an increased number of segments. The combination of these two factors, short 
inter- tuberosities and posterior-sagittal diameters, was essential to reduce seriously 
the efficiency of the outlet. The normal movement of the sacrum, allowing the 
tip to swing backward, enlarged the available space considerably. The diameter 
of the ischial spines, though probably of great importance in the efficiency of 
a pelvis, was not available in this series of ancient pelves. 
4. A numerical variation of the sacral vertebrae was noted in 47 pelves 
or 21'7 °/ u > or one m every five pelves. The number of segments ranged from 
four to six (see Plate IV). By an increased number of segments is meant sacra 
in which the six segments were all sacral in character, or those in which there 
was a transitional vertebra, whether lumbar or coccygeal in character. Such a 
classification, as was pointed out to me by the late Professor T. Dwight, is more 
practically useful than anatomically correct. Pelves with small outlets were 
found slightly more often among those with an increased number, and it is 
possible that this increase in segments may be one small factor in reducing the 
size of the outlet. Aside from the possibility of slightly infringing on the outlet 
space, in rare cases the numerical variation appears to play no important part in 
the variation in size and shape of the pelvic cavity (Plate VII, 9 and 10). 
5. False and double promontories were found in twenty pelves, about one 
in every eleven pelves. These false promontories varied from a marked prominence 
of the second sacral vertebra (Plate VI, 6), an equal prominence of the first and 
second sacral segments, to a projection of the top of the last lumbar vertebra 
beyond the sacrum nearer to the pubes (Plate V, 5). In nearly all cases false 
promontories were associated with transitional vertebrae or an increased number 
of sacral segments. The apex of the lumbo-sacral bend falls at a point pro- 
portionally distant, in all probability, from the sacro-iliac attachment or the 
" vertebra fulcralis " of Welcker (13), and in these transitional cases this distance 
brings the point on another vertebra than the usual one. False promontories 
occurred in four specimens classed as moderately generally contracted and in one 
with a small outlet, but in no other " contracted " pelvis. That is a little more 
often than the general average, but in all probability has no special significance. 
From the general appearances as well as from the measui-ements it seems fair 
to say that the false promontory has no appreciable effect on the pelvic cavity. 
