696 E. KIRKrATKICK. 



whetlier right or wrong, is based on tlie prolonged investiga- 

 tion of abundance of good material, whereas other opinions 

 seem to me to be founded mainly on a priori considerations. 



I have examined over 500 specimens of Merlia and have 

 always found the tissues of the sponge iu most intimate 

 association ^ Avith a calcareous structure, which grows as the 

 sponge gi'ows. Grranular amoebocytes varying in shape 

 according to circumstances are found everywhere in contact 

 with the skeleton, ;uk1 apparently in continuity with the 

 ■organic matrix of the same. Also there is a curious 

 similarity in appearance between the granules of these super- 

 ficially })laced cells and the surface view of the ends of the 

 fibrilhe of the skeleton. The amoebocytes in the upj)er ])art 

 ■of the sponge have the same fundamental characters as those 

 in the interior of the crypts. 



Amoebocytes are found deep dowm in crj'pts nearly closed 

 over by tabulas, and it is incredible that these large masses 

 of cells could have worked their way down through the 

 almost closed slits Avhich are found in many tabult^. Apart 

 from these mechanical difficulties, one cannot imagine, from 

 the point of view of common-sense, why the under surface of 

 a supposed pai-asitic sponge should send down cylindrical 

 moniliform masses of granular cells into the empty s])aces 

 ■of a foreign organism, thereby carrying out a seeminglv use- 

 less and exhausting procedure. In young sponges on delicate 

 shells, well stained and perfectly trans])arent, it can be 

 clearly seen that there is not the least trace of any other 

 organism than the sponge (on the shell). ^Veltner writes — 

 iind not unnaturally — of the calcareous structure as that of an 

 unknown organism in which a sponge has settled. If this be 

 so the said organism has preserved its incognito in a uuirvellous 

 nuuiner. The theory that Merlia is a sponge that has formed 

 both a siliceous and a calcareous skeleton seems to me the only 

 one ])ossible, and further^ the theory that it is a siliceous 



' Of course, it is usual for a parasite to be closely associated with its 

 host, but I trust that this investigation wiU make it clear that Merlia 

 is not an instance of an association of tliis nature. 



