Coi'relatioa of Brain Cupatltij and I nlcliijcnce. 259 



liviiijj: suhjoct. and upon men of tlir same nationality as the vast 

 majority of those lierein ilealt with. We very naturally felt, there- 

 fore, thai in rstal)lisliiii<,' a diriTt coinparisoii hctwueii I he lengths, 

 breadths and heights as ;_Mven i)y (iriHiths foi- living murderers 

 with the saniL' mcasuicmciits as riH-oi(l(Ml by ourselves upon the 

 «;ists. \vx' shoidd be in an infinitely strong(!r jiosilioii to make the 

 ♦■orreet ions wiiich the evidence at our disjiosal showed Mere neces- 

 sary for 1 he Mi'lbourne criminals. As a control experiment we also 

 incorporated in the i-om{)arisoii the lengths, breadtlis. and heights 

 of I I criminals jtreviously recorded by us (Ij. These last measure- 

 ments were all made ujxin thi; living subject, and comprise 11 

 imiles i-onvicled for the ciime (d' murder, with the sentence of death 

 subse(piently commuted to penal servitude. 



The comparisons of the lengths, bi-eadths ami heiuhts fo)' ihese 

 three gi-oups is set foith ill TabU' 11. In every instance the true 

 means foi- lengths, bicadths and heights is gi-eater for the .\lel- 

 boui-ne cases than foi- either the "it) Paikhurst living criminals or 

 the 11 living .\lell)ourne mui-dcrers. an(ither pro(d", if indeed fur- 

 tiier ]iroof lie re(piired. of the necessity of making some correction 

 for ciiir ■^'> condemned Melbourne murderers. Omitting decimal 

 poinls. the true means of the lengths of heads as measured upon 

 the casts e.veeed the Parkhurst figures by 7 mm., and the small 

 group (d' 11 reprieved Melbourne mtirdereis by in nini. The cor- 

 responding excess for the breadth is 10 and H mm., and foi' height 

 2 ami 4 mm. rcs|)ectivt'ly. 



On e.xamining the data afforded by this table it is at once clear 

 that of the standard gioiips of com])arison, the small .Melbourne 

 group is open to the objectimi of insutiiciency of nvnnbers, and 

 also affords the greater ])ossil)li' errors throughout. Further, had 

 we correcte(l the Melixuiiiie casts from the figures attained l)y the 

 11 living and reprieved Melbourne murderers, the i-ubic capacity 

 of Inain for the 3.} murderers hangeil would have hvrw I'tmsiderably 

 less than we sliall subse(|Uently show it to be. 



(hi the other hand if we establish direct comparisons between 

 the true means of length, breadth and height for the .■)(*) Parkhurst 

 <riiiiinals. and the •)•"» .Melbourne casts, we have a I'lose a])proxima- 

 tion <if numbers, ami an excess of the former over the latter of, 

 for length. 7 mm., breadth lO mm., and height 2 mm. If, there- 

 fore. We were to correct the Melbourne figures as recorded upon the 

 casts by de<lucting therefrom the figui-es just mentioned, we should 

 establish a, correction which would conform, on medical grotmds. 

 in the m<ist singularly exact way. with what we should expect to 

 find as the results of oedema. It is clear that any oedema w'ould 



