The Development of Ischnochiton. 629 



It is with comparative difficulty that the conditions are made 

 favorable for a further development of the larvae. If they are placed 

 on ulva the greatest care must be taken to change it frequently and 

 even then the young soon commence to develop abnormally. I have 

 found that somewhat water-worn shells of Mytilus afiPord the best 

 material for the larvae to rest upon. In this case however the shells 

 must be replaced at least every two days by others, the debris that 

 collect about the young must be carefully removed with a camel's 

 hair brush, and also in case the sea water is at all filled with sediment 

 organic or inorganic it is best to pass it through a filter before 

 allowing it to run into the aquarium. 



At first the larvae half crawl and half drive themselves along 

 by means of their cilia and frequently leaving the object upon which 

 they rest they swim about again. In such cases it appears that the 

 resting place was unfavorable, at all events it is readily seen that some 

 places on the shell or ulva are closely packed with settled larvae 

 while other equally accessible portions are nearly or quite bare. 



For some time after the larvae settle they remain quite active, 

 contrasting strongly with the slow steady moving embryos a day or 

 two older. If disturbed or irritated they will sometimes bend the 

 head vesicle either to the right or left or dorsally almost at right 

 angles to the trunk. This occurs until the shell has commenced to 

 form but after the portions of the valve have fused into continuous 

 plates such movements entirely cease. 



2. Metamorphosis. 



As may be readily seen, the metamorphosis of the free-swimming 

 larva in assuming the adult characters is relatively slight and concerns 

 almost altogether the head vesicle. At first this is almost hemispherical, 

 but later it becomes pear-shaped or conical with the tuft of flagella 

 at its narrower end. 



The shell occupies a portion of its upper surface and in its sub- 

 sequent changes this region undergoes little change of shape while 

 the remaining portions of the head vesicle undergo modifications which 

 result in the formation of the anterior portion of the mantle and the 

 proboscis. 



A day after the larva has left the egg membrane the anterior 

 hemisphere shows no very decided permanent change of form although 

 it is constantly changing its shape and exhibits a slight tendency to 

 become more blunt. This latter feature is probably aided by the first 



