CORBITELLA AND HETEROTELLA. lo 



gradually and without any abrupt change in character into those 

 of the lateral parietes. 



According to Marshall the sponge consists of unfused 

 spicules, which statement, as he himself gave it to be understood, 

 is based on but a hasty observation with the naked eye. To me 

 it at once seemed apparent that all the main skeletal beams — 

 those of the sieve-plate included — consisted of spicules soldered 

 together in the usual manner. The truth of this observation was 

 later confirmed by microscopic examination of a beam fragment 

 as well as by a note received from Professor Max Weber, who, 

 at my request, kindly determined the fused state of elements 

 in the sieve-plate beams. Spicular fusion existing in this struct- 

 ure, there can be no question as to the same condition obtaining 

 in the skeletal lattice-work of the lateral wall ; for, it is a 

 recognized fact that the soldering together of spicules begins 

 at the base and proceeds continuously towards the apex of the 

 sponge-body. 



Of the loose tissue half a dozen preparations stand at my 

 disposal for study. So far as are represented in these, the sup- 

 porting parenchymalia are mainly diactins (length 4 mm. and 

 under ; breadth near the center 6-45 fj) and much less frequently 

 hexactins with slender rays and of variable dimensions under a 

 medium size. There also occur at intervals the same slender 

 microxyhexactins, in both the smooth and the peculiarly spiny 

 variety, that were found in Gorbitella speciosa. Axial length of 

 microxyhexactins 330-150 p- ; breadth of ray near the center 3 or 

 4 /a Scattered here and there are the small, plump-looking, 2-6 

 rayed spicules which had been described and figured by Marshall 

 ('75> %s. 60 e-h) and to which I have already alluded under 

 Corbitella speciosa (p. 8). Axial length 57/^ and over; thickness 



