46 I. IJIMA : HEXACTINELLIDA. I. 



The dermalia — which term I adopt in the sense of s^Dicules 

 arranged in a hiyer or layers in the ectosome and at the most 

 peripheral position in relation to all other parts of the skeleton, 

 not in the narrower sense of spicules belonging to the so-called 

 dermal membrane — are always sword-shaped hexactins (PI. IV, 

 fig. 28 ; PI. V, fig. 36 ; &c.). The greatly prolonged, proximal 

 blade-ray penetrates the choanosome like a nail and materially 

 contributes to the firmness of the latter. The distal hilt-ray is 

 the shortest, but is not otherwise strikingly distinguished from 

 the rest of the rays. The paratangential guard-rays of se^^arate 

 dermalia form at places a tolerably regularly quadrate-meshed 

 latticework. This lies a short distance below the external 

 bounding surface which is lifted up by each hilt-ray in a tent- 

 like manner. This position of the dermalia would account F. E. 

 Schülze's constantly calling them the hypodermalia. To me it 

 seems that this ' hypodermal ' situation is simply due to the 

 presence of the distal rays ; were these to atrophy, the trabeculai 

 outside the paratangential layer would lose their support and 

 cease to exist, whereby the said layer would be brought to a 

 position as superficial as the * autodermalia ' of certain other 

 families (Caulophacidse, Rössel lidse). Plence, I have considered 

 it more in conformity to the circumstances to call the single- 

 layered spicules in question of ^wj9/ec^e//iV?œ simply the dermalia, 

 reserving the terms autodermalia and hypodermia for use only 

 in the cases, in which the dermalia are differentiated into an 

 outer and an inner layer respectively. 



In Euplectella, the hexactin-dermalia along the edge of the 

 cuff and of certain parietal ledges may in some species be con- 

 siderably larger and stronger than those on the general surface. 

 And, frequently some of these large dermalia are seen situated 



