14 Miss 13. B. Crozier on Bermudian 



Several entire buds of various sizes, after having been 

 treated as described above, have been examined and found to 

 contain the same four classes of spicules which occur in the 

 adult sponge. The structure of the various filaments sent 

 out by this sponge was studied from teased living material. 

 The anchoring filaments, bud-stalks, and fibres projecting 

 from buds are composed of a core of megascleres surrounded 

 by a dense layer of amcebocytes, among which are scattered 

 numerous chiasters. A \'ew spherasters are found, especially 

 in the little swellings at the ends of filaments. 



Donatio, seychellensis, first observed in Milbrook and 

 Fairyland Creeks the first week in September, became more 

 and more abundant, and the buds more numerous, until 

 about the end of the month. By the middle of October 

 there was a marked falling off in, numbers, and those found 

 were small and with few buds. During the winter only 

 occasional specimens could be found. A recent collection, 

 made on May 14, was a little more successful ; it resulted in 

 eight small specimens without buds. 



Search has been made for D. seychellensis, as well as for 

 other species here described, in most of the common shore 

 collecting-grounds about Bermuda. In only one locality, 

 besides the creeks mentioned above, has D. seychellensis been 

 obtained, viz., on the shore of Agar's Island, where at low- 

 tide level a few specimens have been found attached to stones. 

 A careful search with a water-glass upon the reefs at Daniel's 

 Head Flat failed to reveal any species of Donatio,, as have 

 also dredgings in Great Sound. 



II. Donatio ingalli (Bowerbank). 



The two sponges here classified as varieties of D. ingalli 

 do not correspond in all points with previous descriptions 

 of that species ; but in view of the fact that these descriptions 

 have generally been based upon small numbers of preserved 

 specimens, with no data as to the variations displayed by any 

 one species in its natural habitat throughout the different 

 phases of its reproductive cycle, it seems to me that nothing 

 short of differences in the more fundamental points of internal 

 anatomy warrants the recognition of separate species ; and 

 between Donatio ingalli as previously described and the 

 Bermuda varieties there are no differences of this nature. 

 On the other hand, in cases where both the animals in 

 question are available in the living condition in unlimited 

 numbers, superficial characters, if constant, are a sufficient 

 reason for separating species; on these grounds there can be 



