162 
r.EOFFREY SMITH. 
Table I. 
Size in uim. x 5.A0. 
5. 
0 . 
7. 
8. 
9. 
10. 
11. 
12. 
13. 
11. 
15. 
IG. 
Totals 
Normal pigeon 1 
6 
150 
25 
181 
Normal pigeon 2 
6 
70 
40 
8 
124 
Normal pigeon 3 
1 
38 
50 
10 
99 
Normal dove 
1 
23 
51 
24 
3 
102 
Hybrid A . 
1 
6 
y 
6 
1 
5 
7 
13 
19 
17 
3 
1 
88 
Hybrid B . 
— 
— 
— 
— 
4 
5 
9 
25 
29 
5 
2 
— 
79 
Hybrid C . 
1 
2 
3 
2 
2 
5 
6 
14 
32 
13 
3 
1 
84 
Sum of hybrids . 
2 
8 
12 
8 
7 
15 
22 
52 
80 
35 
8 
2 
251 
Sum of normals . 
14 
281 
166 
42 
3 
— 
— 
506 
The taV>le shows the frequency distribution of sizes of random 
samples of spermatozoa heads from normal and hybrid birds. The 
modal value for the normals lies between 6 and 7 ; for the hybrids, 
between 12 and 13. 
It will be seen tliat wliei’eas tlie spermatozoa of the normal 
males vary from 5-9 (to get the actual measurement in mm. 
these figures must be divided by 550), with the modal 
value between 6 and 7, the spermatozoa of the hybrids 
vary from 5-16 with the modal value between 12 and 
13 ; in other words, the majority of the hybrid sperma- 
tozoa are just twice the size of the normal. In the last 
two horizontal columns the frequency of the various sizes 
of all the hybrids summed together and of all the normals 
is given, and if Ave regard all the hybrid spermatozoa 
measuring from 11-16 as being double the normal size, then 
79 per cent, of the hybrid spermatozoa may be regarded as 
double the normal size. This is probably an under-estimate, 
owing to the reason already given, viz. that the beaded and 
twisted spermatozoa which are not included are practically 
all of the double size, and they constitute about 50 per cent, 
of the total number. 
Having established this curious abnormality of the hybrid 
spermatozoa, in point of size, the next step avas to inquire 
Avhether any explanation of it could be found in the pro- 
cesses of spermatogenesis in the hybrid testes. 
