246 DE. N. ANNANDALE ON FKESHWATER 



cup-shajDed chitinoid funnel. The spicules on the gemmule are arranged 

 in two layers, one o£ which lies parallel to the external surface of the 

 gemmule outside, while the other is tangential to the internal coat and 

 is more or less irregularly arranged ; in form the gemmule-spicules resemble 

 the flesh-spicules but are always stouter. 



Of Scottish examples of this sponge I possess three sets, — one from the 

 Glasgow Canal at Edinburgh, given me by Mr. W. Evans and taken in 

 October 1907, one (a single specimen) given me by Sir John Murray and 

 labelled " River Dee at Crossmichael, water 4 to 6 feet deep, 27th June, 

 1905," and one obtained by myself from Loch Baa in Mull, in October 

 1907. Each of these sets of specimens represents a different form or 

 phase of the species ; but it is, in my opinion, better not to separate them 

 as varieties. The term '^ var.," at any rate in the Spongillinse, has 

 several different meanings, such as local race, temporary or seasonal phase, 

 modification directly due to environment, distinct and permanent form 

 distinguished by small but constant differences, and so on : in most cases 

 it is impossible to say exactly what it means, and its use is merely the 

 confession of a desire " to beg the question." It seems to me that if two 

 forms are found growing together on a number of occasions in a number 

 of localities, and if the specimens taken on different occasions are con- 

 sistently different inter se, they should be regarded as distinct species, 

 no matter how small the difference may be. If, however, different forms 

 are only found in different localities or environments, or at different times 

 of year, but agree in the majority of their characters, then they are probably 

 no more than temporary phases which would not breed true if their 

 environments were changed. It must be confessed, however, that such 

 rules are easier to make than to keep. Especially in the Eastern tropics, 

 the Spongillinse have only been studied in a very few localities ; I have 

 rarely visited a new locality in India without finding forms that were 

 previously unknown to me and did not agree with any published description. 

 It has not always been possible to decide in a very definite manner whether 

 these were true species or not, and I have usually adopted the system of 

 giving even doubtful species names, trusting to future researches to confirm 

 or disprove their valency. It is almost impossible to make progress 

 in classifying or arranging a large collection unless the specimens are 

 given names. Personally I regard many of the species I name without 

 seeing " types," whether they be " new " or old, as named provisionally. 

 Many published descriptions of the lower invertebrates are quite inadequate, 

 and organisms so plastic as the sponges, as a matter of fact, conform 

 imperfectly to any system of nomenclature as yet defined ; unfortunately 

 it is almost impossible, at any rate in a warm climate, to keep them alive 

 and healthy in captivity, and so study their variation from generation to 



