512 HYATT'S REVISION OF THE 



first mentioned refers to a specimen of sub-S23ecies tuhuUfera, and the second to the corres- 

 ponding quahty of sub-species Mediterrcmea. It will be observed that the divergence is 

 greatest both in form and texture in the finest quality, figs. 15 and 21, and that the 

 approximation of the varieties occurs in the lower or less regular and coarser fonns, such 

 as figs. 10 and 20, and still more closely in figs. 16 and 18. The first of these last is a 

 fragment of an old specimen of tuhuUfera, and the last is the surface of a very coarse 

 specimen of 3fediterrcmea. 



There are several prominent variations in the mode of arrangement of the fibres when 

 examined by the microscope, of which the following are perhaps the most remarkable. 



1. The primary fibres single, continuous, projecting, stout. From the teruunations of 

 these hang festoons of secondary fibres, forming the superficial, interstitial skeleton. 



2. The primary fibres either single or treble, and more or less continuous below as a 

 sort of closely woven fascicle. 



3. The primary fibres project in a fascicle of two, three or more, but are deflected 

 latei'ally below, and ai'e not continuous in the body of the skeleton as vertical fibres. 

 Their presence is indicated below by a denser mesh of fibres surrounding a continuous, or 

 nearl}^ continuous, and somewhat tortuous track of debris. The mesh of connecting or 

 secondary fibres is proportionally about equal in length and breadth in all the specimens 

 examined. 



This sub-species has, as compared with other American sub-species, a skeleton composed 

 of remarkably fine fibres, which bleach out to a whitish brown color. The surface is covered 

 with fine tufts of primary fibres which are, however, very pliable. The surface is generally 

 quite free from cushions and ridges, and the channels between these when they do occur, 

 are neither very deep nor long. The result of these characteristics is a form with a 

 smoother surfiice and a denser looking skeleton than usual, pierced on the sides by numer- 

 ous small apertures very regularly distributed, and at the top by one or more large cloacal 

 oscules. The form is generally dome-like, and is never, so far as I know, cup-shaped, 

 though it may become exceedingly irregular, fistular, or even dendritic. 



The older specimens show a decided tendency to increase by the prolongation of the 

 parts immediately around the apertures. Thus the main body of the sponge becomes pro- 

 jected into numerous smaller conical or head-shaped masses, like the young of variety 

 rotunda, and crested masses like those of variety dlsciforinis. At an advanced age the 

 fibre becomes very brittle, and unfit for domestic purposes. 



Dr. Palmer reports that this species is found generally associated in considerable numbers, 

 not often growing either upon other sponges or upon corals, but generally upon hard bot- 

 tom. Color while living is black, and the largest seen was about eight inches high by 

 about twenty in circumference. 



Loc, Key West, Nassau. Habitat, hard bottom or reef in six feet of water. 



Variety ^er^Msa (PI. xvi, fig. 11-lla). The colony in this variety is composed of several 

 fistular elevated orifices of the ordinary density of texture, but united by a depressed por- 

 tion of extremely porous structure. The secondary tufts or pencils are of extraordinary 

 length upon the flattened central portions. Under this head is included a number of forms 

 in which the fistular orifices are numerous. These being more closely approximated than 

 in other sub-varieties, this intermediate depressed portion disappears, and the sponge 

 acquires the characteristics of the next variety. 



