544 HYATT'S REVISION OF THE 



the skeleton, and the accumulation of foreign materials, fine particles of sand, foi'aminiferte 

 and broken spicules, etc. When held up between the observer and the light, numerous 

 small canals i-adiating and branching from the base reticulate the interior. These are pen- 

 etrated by numerous closely adjoined incurrent pores. The excurrent pores are all on the 

 upper side and of various sizes, though none very large. The accumulation of foreign 

 matter is also much less on the lower side, so that the reticulated aspect of the surface can 

 be observed. The longitudinal thi'eads are radiatory, but are irregular in size and shape, 

 and loaded with spicules and foreign material. 



The connecting or secondary set of fibres are not placed at right angles to the primary, 

 but are very irregular. They do not differ greatly in size, and are so slightly colored as 

 to be almost perfectly diaphanous when mounted in balsam. They are also nearly com- 

 pletely free from foreign matter. Besides these, there is still another system of extremely 

 minute fibres connecting both of the former irregularly with a fine network similar to that 

 found in Stelospongos, but diflfering m its want of regularity and the manner in which it 

 permeates the whole structure. These characteristics are wholly Spongian, and do not 

 seem to confirm Ehler's conclusion that it is a form closely allied to Chalina, though they 

 justify his description of the form as a distinct genus. Ehler considers that Esper's Sp. pli- 

 cata shows considerable affinity for this group in its charactei'istics, but of this I have never 

 seen an example, or even Esper's figure. 



Loc, Cape of Good Hope, in Ward's Museum and in Soc. Coll. 



HIRCINIAD^. 



The great contrast presented by the aspect of the skeleton in this family as compared 

 with that of either the Spongiadte or Phyllospongiadae, would justify a separation inde- 

 pendently of more minute structural difierences. There is not that distinction between 

 primary and secondary fibres which is to be found in the two preceding families. When 

 debris is present in the water the fragments, which must have been derived from the dermis, 

 are not confined to the core of the primary fibres alone, but are equally ^Ji'evalent in the 

 secondary fibres, showing that they must have been formed by the same layer of the der- 

 mis as the primai'y fibres. Both classes of fibres, also, are sarcodous or gelatinous, and are 

 readily dissolved in some specimens by maceration in cold water. They are also flatter in 

 form, and have a fibrous aspect when dried, in some species due probably to this flattening. 



Dysidea approximates most closely in the color and aspect of the fibres to those of 

 Spongia, but can never be mistaken for a true Spongian, though the fibres evidently pos- 

 sess more keratode than those of any other species in the family. The forms of the speci- 

 mens are similar to those prevalent among the Spongiadse, dendritic, solid, flabellate, fistular 

 and cup-shaped, but never frondose, as in the Phyllospongiadce. 



Dysidea Johnst. 



Spongelia (pars) Schmidt and Bowerbank. 



The skeleton in this genus, as in the single species which appears to represent it in Flor- 

 ida, appears to combine the characteristics of several genera. The primary fibres are large 

 and sometimes partially compound, as in some species of Stelospongos ; the secondary 

 fibres partly like those of Spongelia and partly like those of Hircinia. Not only are the 

 primary fibres loaded with foreign material, as in Spongelia,, but the secondary, also, and 



