333 
1887.] Mr T. B. Sprague on a Fruitful Marriage. 
some cases we have also particulars as*to the grandfather (being the 
father’s father), the father’s brothers, and their sons, who are 
respectively uncles and cousins to the peer. The existing peer at any 
time may thus he regarded as the central figure of a group, around 
whom are arranged his different relations, at a greater or less dis- 
tance, according as their degree of relationship is more or less remote. 
Assuming now that the precautions we have taken have secured 
that all the facts we extract from the Peerages are equally trust- 
worthy, we have next to consider whether they are all equally suit- 
able for our purpose. Our object is to obtain particulars of a large 
number of marriages, which may be considered as a fair sample of 
the whole; and then to ascertain how many of these resulted in the 
birth of issue, and how many were childless. If, then, we extract all 
the marriages of the peer and of his above-mentioned relations, will 
these give us a fair average of cases suitable for the solution of our 
problem 1 ? Or is there anything in the principle on which our 
selection of facts is made, that renders the marriages we select 
unsuitable representatives <sf the general body *? One such circum- 
stance becomes obvious at first sight. The peer may be single, or 
married; and if the latter, he may either have children, or have 
none. The same is the case with regard to his sons, his grandsons, 
his brothers, his nephews, his uncles, and his cousins ; but we see 
that his father and his grandfather had at least one son each ; each 
of them, in fact, being included in our list for the very reason that 
he had a son. It follows that, if we include in our enquiry the 
information regarding the fathers and the grandfathers of the peers, 
we shall not obtain trustworthy results ; for we shall have an undue 
proportion of fruitful marriages. If we include in our list the 
marriages of the peers’ fathers, we must, in order to get the proper 
proportion of childless marriages, include also the marriages of all 
their contemporaries who were married and had no children ; but to 
attempt this would, I believe, be an impracticable task. If we took 
only those fathers of peers who were themselves peers, we might 
perhaps obtain a suitable body of statistics, by taking the marriages 
of all contemporary peers who married but had no children. In 
some cases these were succeeded by their brothers, or nephews, or 
other relatives ; and all cases of this kind could probably be 
traced without any great difficulty ; but, in a number of other cases, 
