376 DISCOVERY REPORTS 



external form is a more reliable feature than the skeleton ; that it is more constant than 

 the shape of the spicules, and far less variable, under normal conditions, than is usually 

 assumed.^ In addition far too little attention has been paid to the construction of the 

 skeleton. In my opinion, the surest guide for the diagnosis of families is found in the 

 types of spicules present ; for genera, their arrangement and disposition in the skeleton, 

 with minor variations in their shape; for species, the external form, with the minor 

 variations in the arrangement of the spicules, and, less important, the categories of 

 spicules present. In other words, the best method of identification lies between that 

 adopted prior to 1860-70 and that which has been in use since. 



Another practice which is to be deplored is that of founding species on badly damaged 

 specimens, fragments of specimens and obviously pathological individuals (including 

 those which occasionally replace their proper skeletons by sand or foreign spicules). 

 Such objects, used as holotypes, give infinite trouble and rarely can their true characters 

 be ascertained. 



If, however, the skeleton is to be used in so important a role, it is at least time that 

 some systematic effort were made to determine the extent of variation (i) in size, (ii) in 

 shape, and (iii) in arrangement. Also, to collect data on the extent to which categories 

 of spicules may disappear in individuals of a species, or be differentiated into two or more 

 sizes. In studying the sponges collected by the Discovery expeditions, the following 

 observations have been made and appear worthy of record : 



Variation in the size of spicules 



The first important observation on this point concerns Calyx arciiarius (Topsent). The 

 oxea in the type measure 0-35 mm. long, in the specimens recorded by me (1929, p. 422), 

 0-27-0-35 mm., not 0-2-0-3 mm. as stated, in the present specimens of the Discovery, 

 0-22-0-24 mm., and in a specimen collected at South Georgia by Mr P. Stammwitz, 

 0-15 mm. long. The spicules of the last specimen are less than half the size of the maxi- 

 mum recorded. 



In Axociella flabellata (Topsent) the large styli and dermal subtylostyli have a maxi- 

 mum of i-o and 0-650 mm. respectively, in the Terra Nova specimens and, in one of 

 the present specimens, of i-2 and 0-640 mm. In the other Discovery specimen and 

 a specimen collected by the Shackleton-Rowett expedition, 0-64 and 0-35 mm. respec- 

 tively. Again, the maxima recorded are twice the recorded minima. 



In lophon proximum (Ridley), the acanthostyli vary from 0-14 to 0-44 mm. (taking 

 average sizes in different individuals), and the tornota from 0-14 to 0-315 mm. 



It is clear from these examples alone that such phrases as " this species differs from . . . 

 in the greater size of its spicules", are useless. Ceteris paribus, spicule size can have no 



1 Strong objection is here taken to the all too prevalent use of the term "amorphous" in describing the 

 external form of sponges. The term has been used to include encrusting, massive, lobate, sub-digitate, 

 clathrate and plicate sponges, and a number of combinations of two or more of these. The type of one of 

 Lendenfeld's species of Australian Chalininae is a beach-worn sponge which has been, presumably, trodden 

 under foot. This was described as "amorphous" and is the only truly amorphous sponge I have ever seen. 



