D. H. DB Souza 
495 
Of the measurements which can be taken on the living subject, next to the 
diagonal conjugate the height of the hip bone shows the closest correlation with 
the obstetric conjugate (right - 55, left '56). Equations were formed using the 
height of each hip bone 
(7= -590^-1-037 + -619 (g), 
C=-602L -1310 + -610 (h). 
The equations as tested by probable errors are better than (b) and (e) but not 
as good as (a), (c), (d), and (/), but if each of the measurements be taken with the 
diagonal conjugate the resulting equations are exceedingly good 
G = 1-008X) - -049 P - 977 + "298 (i), 
G = 1-005 D - -(U2L - 1-075 + -298 (j). 
The probable errors to four places would be '2976 and -2978 respectively while 
that of (c) is '2984, so that the probable errors are less than that of any equation 
except (/). These equations have a practical advantage over (/), for the height 
of the hip bone can be easily measured on the living subject whilst, as already 
mentioned, there are disadvantages attached to the measurement of the pubic 
height. Moreover, the measurement of the height of the hip bone is less 
objectionable to the patient than that of the antero-posterior diameter, so that 
equations (*) and (j) possess a practical advantage over (c). 
To facilitate reference all the equations for finding the obstetric conjugate are 
now collected together: 
(a) 
G = 
■978I>- 1-553 + -299. 
(b) 
C = 
•365^+6-081 + -705. 
(o) 
G = 
■972D + -023^1 - 1-743 + -298. 
(d) 
G = 
•972D + -0234 + -0017- 1-745 + -299. 
(e) 
G = 
•HOP + 10011 +-738. 
(/) 
C = 
•989Z>--165P- 1-229 + "293. 
(9) 
G = 
•590P- 1-037 ±-619. 
(h) 
G = 
•602X- 1-310 ±-610. 
(?) 
G = 
1-008P- -049P - -977 ± -298. 
(j) 
C = 
1-005P- -042X- 1-075 + -298. 
Equations (/), (i) and (j) appear to indicate that, for a constant diagonal 
conjugate, the obstetric conjugate is smaller the higher the pelvic wall. 
These formulae have been tested for the original series and for five other 
pelves. The cards of 25 of the 216 pelves were picked out at random and the 
obstetric conjugate was calculated from the recorded measurements by means 
of the equations, with satisfactory results. The square root of the mean square 
deviation of the calculated from the recorded values of the obstetric conjugate was 
multiplied by - 67449 and compared with the probable errors of the equations. 
The values calculated from equations (b), (g), and (h) fell within this probable 
63—2 
