150 Mr Heape , Notes on the Proportion of the Sexes in Dogs. 
pups among Greyhounds under the conditions of breeding now 
practised, and this result is attributed to a selective action on the 
ova produced at this time. 
Collies. The above returns are compared with similar ones 
obtained for Collies (Table VI). The latter show much less 
variation in the proportion of litters produced at different times of 
the year, the breeding of these animals being conducted much 
more in accordance with the natural breeding time for the species. 
Still in Collies there is a marked increase of births from May to 
August, and from October to December is the season when fewest 
births occur. 
This result is due to the number of litters produced and not 
to any variation in the rate of fertility per litter. 
The returns for Collies show no evidence that conception at 
any particular time of year affects the proportion of the sexes 
born ; they rather indicate, if my supposition is correct, that when 
the animals breed at a time which is unfavourable for the rearing 
of the pups, special care is exercised. 
Kennel Register and Breeders Schedules. Data obtained from 
the Kennel Register for the years 1889 — 1891 (Table IV) are 
checked by information derived from Schedules filled up by breeders, 
in which latter returns accurate details are supplied for all the 
litters there dealt with (Table VII). In both Tables details are 
given for each breed dealt with and these are again classified 
under headings of Large dogs and Terriers. It is found that the 
fertility per litter is the same for Terriers in both the Kennel 
Register and the Schedules, while the latter show '57 smaller 
litters for Large dogs. On the whole it may be assumed that 
the Kennel Register very generally includes all pups born. Where 
this is not the case it would appear that still-born pups or pups 
destroyed on account of weakness, which are not recorded in the 
Kennel Register, are usually of the male sex, since there is a 
considerably higher proportion of dog pups recorded in the 
Schedules than in the Kennel Register. 
It is highly probable that the animals included in the Schedule 
returns are more in-bred than are those entered in the Kennel 
Register, and that this fact may account for some of the excess of 
male births recorded in the former. There is not wanting evidence 
from other sources that in-breeding is associated with the produc- 
tion of a high proportion of males. The bulk of' the data at my 
disposal for dogs is not of a character which admits of wide 
generalisation on this point ; there is, however, some evidence in 
the Schedules that an excessive production of males is associated 
with in-breeding and the details given of Bloodhounds may be said 
to be strongly in favour of this view. 
In both these returns the Large dogs are found to produce 
