EVOLUTION OF SCIENTIFIC SERIALS. 223 
in original scientific investigation, many of them have other 
means of publication ; and while their communications are 
read before and discussed by the society, they are frequently 
not offered for printing. Moreover, the growth of the Philo¬ 
sophical Society has always taken place through multiplica¬ 
tion by fission more than through direct increase in stature, 
and the Anthropological Society, the Biological Society, the 
Chemical Society, its own Mathematical Section, and perhaps 
the National Geographic Society, are its offspring. It is 
pertinent to add that the change in form of publication by 
this society was in some degree foreshadowed in the serials 
issued by some of its subordinate branches—the Anthropo¬ 
logical Society recently transformed its organ from a biennial 
volume of Transactions to the quarterly American Anthro¬ 
pologist, and the recently organized National Geographic 
Society sets out with a Bulletin issued at irregular intervals 
as a geographic magazine. 
The Academy of Natural Sciences of Philadelphia was 
selected because it is one of the oldest among the scientific 
societies of the country, because its vicissitudes in member¬ 
ship and financial support have been representative, and 
because its various publications are fairly accessible. The 
record of some of its serials as shown in the conspectus may 
be found defective, (1) because in the absence of original 
covers (removed in binding) it is sometimes impossible to 
ascertain the original form of publication; (2) because in the 
multiplicity of paginations it is sometimes impossible to 
ascertain whether or not certain matter belonged to a given 
volume or series ; (3) because certain of the minor publica¬ 
tions (generally extracted from larger serials) were not found; 
and perhaps for other reasons. 
The Boston Society of Natural History was selected because 
it is one of the older and more important among the scien¬ 
tific societies of the country, because some of its modifica¬ 
tions in plan of publication seem especially significant and 
at the same time representative, and because its publications 
are fairly accessible. 
