334 Proceedings of Royal Society of Edinburgh. [july 4, 
such peers have had no male relations to inherit the title, which 
has therefore become extinct. In order to trace out cases of this 
kind, it would he necessary to have a complete set of old volumes 
of the Peerage, and ascertain by an examination of them which 
titles became extinct each year, in consequence of the peer having 
died without issue, although he was married. From several points 
of view this would be a very interesting enquiry, but it was one 
which I was not in a position to undertake : and if I had desired 
to do so, I cannot see how it would have been possible to decide 
how far back the enquiry should extend, so as to include the 
proper number of holders of extinct titles, but not too many of 
them. In a certain number of cases, moreover, the peers’ fathers 
were not themselves peers, but were untitled men belonging to a 
younger branch of the family ; and I imagine it would be quite im- 
possible to lay down any principle upon which to determine all the 
persons who may fairly, for the present purpose, be considered their 
contemporaries. The difficulty, however, is completely got over by 
excluding from consideration the marriages of the fathers and grand- 
fathers of our peers, or, speaking more correctly, the marriages from 
which our peers were descended. I have, therefore, if a peer’s father 
or grandfather was married twice, or three, or four times, excluded 
from consideration the marriage from which the peer was descended, 
but have made use of the remaining marriages, — subject to a 
correction that will be hereafter explained. I have also, in a few 
cases, rejected all the facts relating to certain peers, when, from very 
exceptional circumstances, such as residence in a foreign country, 
the information was obviously incomplete. 
Another circumstance may be noted : — The peers themselves are 
all alive at the time at which our observations begin, and their 
fathers and grandfathers are all dead ; but their sons, their brothers, 
and their uncles, may be either living or dead. There seems, how- 
ever, no reason for thinking that the fact that all the peers are 
living, will affect to any appreciable extent, if at all, the probability 
of their marriages being fruitful. 
As already mentioned, my investigation commenced with Lodge’s 
Peerage for 1871. The principal object was to ascertain what pro- 
portion of marriages were fruitful : and as many marriages which 
are not fruitful at once, become so in later years, it would be com- 
