1 1 



For e x t e r n a 1 a p pearance and general a n a t o m y , see p. 2 — 6. 



Spicula. 



1. Tylostyle. (PI. IV, fig. 92 — /. ; fig. 13a — £). The tylostyles of/ 5 , carinata may be distinguished 

 into corticalia and parenchymalia. The former occur in the peripheral portion of the grooves, 

 the Iatter radiate in bundies from the axis to the cortex (PI. IV, fig. 2). The paren- 

 chymalia terminate usually (but not always) somewhat obtusely (PI. IY, fig. 9 t, /.), whereas 

 the corticalia terminate in a sharp point. Thiele states the same for "P. mixta." The 

 following table shows the variety in size (in u). 



Xeither Sollas nor Lindgren makes any mention of two sorts of tylostyli, which nevertheless 

 most certainly exist l ). We have taken our measurements from the cortical tylostyli after 

 transverse and longitudinal sections through the grooves. These measurements can be taken 

 without serious difficulty; but the same cannot be said of the parenchymal tylostyles, as 

 these occur in thick bundies. The length of these spicula, we have taken from preparations 

 of isolated spicula. In how far in such cases certain spicula were cortical or parenchymal 

 could not always be ascertained. Hence we have only given the maximum sizes, and an 

 average of the majority. We learn from the above list that the cortical tylostyli vary between 

 146 fi. and 426 (x-, whereas their average in the specimens gives 281 ij.. The average 

 of the parenchymal tylostyles varies between 625 u.. and 1040 u. If then we are told by 

 Sollas that the tylostyles are 1000 u., whereas Lindgren gives 780 ja., we conclude that 

 these authors have measured parenchymal tylostyles only. On the other hand it is probable 

 that the size given by Sollas for the tylostyles of the incrusting P. intcrmcdia (470 u.) is 

 an average taken from a young specimen in which more small (i. e. cortical) tylostyles were 

 present than long (parenchymal) tylostyles, which Iatter prevail in large, branching (adult) 

 specimens. Thiele found for the small (cortical) tylostyles of P. mixta 200 u. for the large 

 (parenchymal) tylostyles 750 y.., a fact which agrees pretty well with our own observations. 



1) Thiele (1900 p. 73) was the first to observe this for his P. mixta; he does not mention, however, the fact that the two 

 sorts occur in different parts of the sponge. 



