496 
The Measurements of the Pelvis 
error. The others exceeded it, but the greatest difference was only "08, i.e., since 
we are working in centimetres, less than 1 millimetre. 
The formulae were also tested for five pelves measured by us. Three of these 
were ancient Egyptian (I and II female, III male), and had to be fitted together 
to be measured, but the other two were modern English (IV female, V male), with 
the ligaments attached. The inter-tuberal diameter was not measured since it 
would have required the application of Emmons's method, and this we were not 
prepared to carry out as it had already been found that, in the determination of 
the obstetric conjugate and the transverse diameter, nothing was gained by the 
use of this diameter. Equation (d) was therefore not tested. The differences of 
the calculated from the recorded values for these pelves are given below for each 
equation. 
TABLE IV. 
Pelvis Number... 
I 
11 
III 
IV 
V 
Equation (a) ... 
1-33 
•39 
- -28 
•15 
- -54 
„ (b) ... 
•83 
- -76 
•52 
2-65 
•33 
„ (c) ... 
1-35 
•40 
- -26 
•19 
- -52 
„ (e) ... 
•44 
-1-09 
•07 
2-00 
- -15 
„ if) - 
1-26 
•59 
-•01 
•28 
- 32 
„ iff) 
1-49 
- -21 
— •51 
2-24 
- 1 -35 
„ (A) ... 
1-43 
- -10 
- -53 
2-42 
-1-43 
„ (i) ... 
1-27 
•37 
- -23 
•07 
- -45 . 
„ U) - 
1-29 
•37 
- -23 
•08 
- -46 
Perhaps the most interesting of these numbers are those for the modem 
English female pelvis IV. Here the results for the equations formed from the 
diagonal conjugate and the height of the hip bone, (i) and (j), are excellent, the 
differences between recorded and calculated values being only "07 and "08 cm. 
The equations formed from the diagonal conjugate singly (a), and the diagonal 
conjugate and antero-posterior diameter (c), are the next best, the differences 
being '15 and "19 cm. respectively, then that from the diagonal conjugate and 
pubic height (/), with a difference "28 cm. All the others have differences 2 cm. 
or more. 
The figures bring out strongly what our previous discussion on the probable 
errors has shown, namely that, excluding, for reasons considered, the equations 
formed with the pubic height, the best equations are those containing the diagonal 
conjugate singly or combined with the height of the hip bone or the antero- 
posterior diameter. 
The equations (i), (j), (a), (c) and (/) give differences less than a centimetre 
in four of the five pelves. No other equation does this except (b), but the 
difference for this in the remaining pelvis, which happens to be the modern 
English female pelvis IV, is 2 - 65 cm., so that the equations which give the best 
results on the whole are those which give the best results with pelvis IV. It 
