J. F. Tocher 
345 
II. Scottish Criminals. The writer is indebted to Dr J. F. Sutherland, 
Assistant Scottish Lunacy Commissioner, for kindly supplying the measurements 
on 375 Scottish habitual criminals. The analysis of these observations has 
provided interesting results. The criminals were divided into four classes ; 
those convicted of (I) robbery and assault, (II) theft, (HI) murder and assaidt, 
and (IV) offences against chastity. The following table gives the results of the 
analysis, stature, iiead-length, and head-breadth being considered. 
TABLE XXVL 
Habitual Cr'iminals in Scotland. 
Mean 
Standard Deviation 
Stature : 
Class No. 
I. 
64 -6 ± 
•11 
2-37 + 
■08 inche 
II. 
65-0 + 
•18 
2-68 + 
•13 „ 
it >5 
III. 
QoS± 
•20 
2^28 + 
•14 „ 
IV. 
65-2 + 
•42 
2 -.56 + 
•30 „ 
Totals 
64-8 + 
•09 
2-47 + 
•06 „ 
Head Length : 
Class No. I. 
„ „ 11. 
„ „ HI. 
» „ IV. 
Totals 
Head Breadth : 
Class No. I. 
» „ n. 
„ „ in. 
„ „ IV. 
Totals 
195 •St 
197- 2± 
198- 4± 
195 - St 
196- 3 + 
•30 
•42 
•63 
•70 
•23 
152- 9± -22 
153^9± ^29 
153- 5 ± -43 
1.52-1 ±1-06 
153-5+ -16 
6^37 ± -21 mm. 
6^06±^29 „ 
7 -06 ±-45 „ 
4-02 ±-47 „ 
6-44+ •le „ 
-70± -16 mm. 
•24 ±-20 „ 
•83±^31 „ 
•50 ±'71 „ 
•57±^11 „ 
Considering briefly in the first place the various classes of crnninals them- 
selves, it is seen that those habitual criminals who have been convicted of murder 
and assault, and in a lesser degree those convicted of theft, differ considerably in 
head-length from those convicted of robbery and assault and other crimes. They 
have on an average longer heads. The difference in head-length between Classes I 
and III is 4'43 times its probable error, and Class III differs in its mean head- 
length from the remaining population by about 3 8 times the probable error of 
the difference. The distinctive feature here is that those convicted of murder and 
assault have significantly longer heads than the other criminal population. The 
difference in head-breadth between Classes I and II, and in stature between 
Classes I and III are possibly significant, but in all other cases the differences in 
the various characters among the criminals are insignificant — they are fair random 
samples of the short series of the general criminal population of Scotland. A 
longer seiies of measurements might of course I'eveal significant differences which 
appear in this series to be insignificant. 
