MR. F. BALFOUR BROWNE ON AQUATIC COLEOPTERA. 2QI 
Reference to my last paper will show that I mapped out 
two curves, one showing the number of species per collection 
per month, of the Hydradephaga or carnivorous Water Beetles, 
the other showing the same of the Paipicornia or herbivorous 
Water Beetles. It was not until I had worked out upon 
the same method the results for 1905 that an error in the 
method occurred to me, and it was owing to the fact that the 
curve obtained for 1905 did not agree with certain observations 
I had made, that I was induced to go more carefully into the 
question of the principle upon which the curve had been 
obtained. 
August, 1905, was a particularly dry month so far as the 
marshes were concerned, although in rainfall it was actually 
wetter than the same month in 1904. I will go into this point 
later on, but for the moment I need merely state that the 
marshes were so parched last August that I only made 3 7 
collections instead of the 100 or so which I had counted upon 
making. I had been paying attention to the question as to 
what happens to the denizens of a pond or dyke when the water 
has all evaporated, and I had worked out from my records 
the fact that the collecting places where the 37 collections were 
made in August were not as a whole richer either in species 
or in individuals than the same collecting places had been 
in June. I was, therefore, surprised after working out upon 
the 1904 method the number of species per collection per 
month, to find that August was much richer than any other 
month during the season. Had I been a mathematician no 
doubt the reason would have been obvious from the first : 
but as it was it took some time before I understood it. The 
error in the method lies in the fact that I had not been con- 
sidering equal numbers of collections in each month, and that, 
under ordinary circumstances, the tendency would be for 
a month in which fewer collections had been made to show 
a higher number of species per collection than a month in 
which more collections had been made. Suppose for instance, 
we have 99 collections, each containing 9 species, and we then 
make one more collection and find 10 species. This additional 
species raises the average for the total number of collections 
