^^Madreporic System''^ of Echinodermata. 325 



has shown that in some Elasipoda whicli have no respirator j 

 trees, the madreporite has retained its primitive position at the 

 surface. A curious converse transformation m,ay be noted in 

 Amphibia, and seems to shed light on the matter. In the 

 aquatic tadpole there are numerous nephrostomes opening 

 into the kidnej, and so on to the exterior. In the adult 

 Anuran, living mostly on land, and consequently absorbing 

 less water by osmosis, the nephrostomes have lost their con- 

 nexion with the kidney and open into the veins. 



I have quoted Sharpey's observation as to the nature of 

 the coelomic liquid, exaggerated by most zoologists into a 

 statement that it was pure sea- water, till P. Geddes redis- 

 covered the corpuscles (" organic particles " of Sharpey). On 

 filtering the coelomic liquid, of EcMaus clear from the clot and 

 boiling it down I obtained a flaky coagulum, which gave 

 Millon's reaction perfectly and which must be due to a dis- 

 solved proteid, only coagulable on boiling, and distinct from 

 the formed elements which compose the spontaneous coag- 

 ulum. 



To summarize : 



1. The madreporic system of Echinodermata is morpho- 

 logically and ontogenetically a (left) nephridium. 



2. Its ciliary current is directed outward through the 

 madreporic disk, and an outward current takes place through 

 the pores of the disk of Comatula. 



3. There is no need for the taking up of sea-water by a 

 perforated plate, since osmosis is amply sufficient for the tur- 

 gescence of dilatable organs. 



4. The rapid contraction or erection of the tube-feet is 

 due to the transference of liquid from one part to another ; in 

 Echinoidea this may be balanced by the protrusion or retrac- 

 tion of the peristome, or by contraction or dilatation of the 

 gut, or in both these ways. 



5. The change of position of the madreporite in most Holo- 

 thuria is probably due to the usurpation of nephridial functions 

 by the respiratory trees attached to the cloaca. 



6. The coelomic liquid of Echinus contains, besides cor- 

 puscles, a dissolved albuminoid, coagulated on boiling. 



I take this opportunity of adding two notes on cognate 

 subjects. 



Note I. — I think it very probable that when an Actinian is 

 at rest the lips of the oral slit are closely appressed, and, in 



