11 
The most serious defects in the work of Potts and his co-workers 
and of more recent workers who have depended on his monograph, have 
been due to the failure to recognize Ephydatia fluviatilis and E. miilleri 
as distinct species. These are two very abundant and widely distributed 
forms and are also highly variable. Specimens of either species were 
indiscriminately designated Meyenia fluviatilis, and nearly a dozen sup- 
posed species and varieties were named and described from various forms 
of one or the other of the two, and later all made synonyms of M. 
fluviatilis, now Ephydatia fluviatilis. In some cases illustrations or 
descriptive statements indicate clearly enough which of the two species 
is involved, but often one can not tell. E. fluviatilis has birotulate gem- 
mule-spicules with shafts longer than the diameter of the rotules, and 
the skeleton spicules uniformly smooth; while E. miilleri has the shafts 
usually shorter than the diameter of the rotules, and the skeleton spicules 
varying from distinctly spined or sparsely micro-spined to nearly or quite 
smooth. Distinctly spined and smooth spicules are often associated in 
the same specimen and in different ratios in different specimens. The 
gemmule walls of E. muiilleri not infrequently include two or even three 
layers of birotulate spicules, while those of EF. fluviatilis uniformly have 
but a single layer. Potts erroneously assumed that the distinction be- 
tween these two species was based merely on the last-mentioned differ- 
ence, and since the appearance of more than a single layer in the former 
species is not a constant character, he decided against the validity of the 
species. 
Another source of error involving North American species, which 
is of less consequence because less confusing, has arisen from the descrip- 
tion by Hilgendorf in 1882 of a Japanese sponge as variety japonica of 
E. fluviatilis. Judging from the description, this is nothing other than 
the form of E. miilleri in which the skeleton spicules are smooth. Among 
the United States National Museum specimens examined by Annandale 
was one from the Potomac River which he listed in 1910 as E. japonica, 
recognizing this form as a distinct species. In 1916 after a study of 
Japanese sponges among which were the various forms of EF. miilleri, 
he treated the japonica form as a variety of E. miilleri. The writer be- 
lieves that there is not sufficient basis for giving it even the rank of a 
variety. It is simply a very common variant form of a highly variable 
species. In the original description of E. miilleri the skeleton spicules 
were described as spined, but by various students of the species in Eng- 
land, Germany, and Austria the extremely variable condition has been 
well known and frequently mentioned. The specimens of this species in 
the writer’s collections from different parts of Michigan, from high moun- 
tain-lakes of Colorado, and from Montana, indicate that those with 
smooth or nearly smooth skeleton-spicules and those with spined spicules 
are similarly distributed ; and also that the same body of water may con- 
tain both of these kinds of specimens, and still others with spined and 
smooth spicules associated in the same skeleton fascicles. Among the 
Colorado specimens, those with spined skeleton-spicules are more frequent 
