174 ENDEAVOUK" SCIENTIFIC KESULTS. 



possession of spined strong via. It is theretore only necessary 

 to assume that the same word in its first occurrence is mis- 

 written tor "oxea," and the description throughout becomes 

 perfectly congruous. In correction of the quoted description 

 of the main skeleton, it is necessary to mention that the main 

 fibres do not, strictly speaking, extend from the base to the 

 margin of the cup ; but as they ascend, trend towards and 

 terminate at the lateral surfaces, whilst new directive fibres 

 branching off from these, continually arise to take their place. 

 In specimens in which the surface is even, the skeleton reticu- 

 lation is very regularly subrectangular, e\ en though the con- 

 necting fibres often form interreticula of small extent between 

 the main ones. With increase of surface irregularity however, 

 the reticulation tends to become confused. The measurements 

 which are given for the fibres and mesh, so long as they are 

 regarded as conveying an approximate idea, may be allowed 

 to stand. It is impossible with no better material than the 

 present, to speak with certainty of the dermal skeleton, though 

 the indications render it fairl} safe to say that anything of a 

 special nature is absent. The errors of previous descriptions 

 in regard to the dimensions of the spicules are partly due to 

 the fact of their great variability in size. The frequency of 

 occurrence of a spicule of given size decreases rapidly as its 

 length recedes from the average, so that one can never make 

 quite sure of the maximum attainable size. This is more 

 particularly true of the principal spicules. Whenever, in any 

 species such great variability in the size of spicules is found in 

 a single given specimen, it is, generall}- speaking, fairly safe 

 to assume that considerable variations in their maximum size 

 are likely to occur in different specimens. I have made such 

 an assumption in the present instance. The largest spicules 

 were found in the type specimens of K. elegans and Thalas- 

 sodeucbon typica, Whitelegge, in both of which the oxea vary- 

 in length from 150 to 620 11 and may attain a diameter of 

 18 \i. In the type-specimen of K. silex, which departs most 

 widely from these in regard to the size of the spicules, the 

 length of the oxea \aries from 130 to 410 \i, whilst the maxi- 

 mum diameter is only 13 \i. Among the slenderest spicules a 

 small proportion of styli are found ; but as these are of equal 

 length with the slender oxea and are frequently curved in the 

 same way, it is difficult to say whether they are to be re- 

 garded as abnormalities or as representing the auxiliary 

 spicules of other species. It is perhaps significant, however^ 

 that they du not attain to more than 5 n in diameter, and 

 that what appear to be transitional forms between them and 

 the oxea are very rare. Their maximum length, which varies 

 in different specimens concomitantly with that of the other 



