61 



together with a rapid rotation. They developed quickly into the 

 form shown in fig. 2, but still remained active. 



After twenty-four hours some were still moving slowly, whilst 

 a few appeared ready to become fixed. Two had not changed 

 in form. On September 8th a small number appeared to have 

 settled down as fixed forms, but on the 10th all were dead. 



Further batches of material were kept and examined from 

 time to time, from which it would appear that the larvae remain 

 active for about twentj^-four hours, after which a further two 

 days gives a complete change from the amphiblastula form to 

 that in which the overgrowth of the granular cells is complete 

 and fixation is taking place (figs. 8 and 9). 



Some of those in the last-mentioned stage appeared nearly 

 twice as large in diameter (viewed from above) as the newly hatched 

 ones. 



The freshly hatched larvae swim at the surface of the water 

 except when disturbed, when they seek the lower layers for a 

 time, but as proliferation of the granular cells increases the larvae 

 sink to the bottom, Avhile still retaining their motile character 

 for a short time. 



In one of the earliest samples an embryo was found to have 

 hatched prematurely, being still in the typical pseudogastrula 

 stage. It had not been subjected to centrifuging. 



The latest batches examined (September 10th) mdicated 

 that the period of hatching was drawing to a close, as in one case 

 eight sponges were centrifuged, and gave only about half-a-dozen 

 larvae. 



An attempt was made to rear the larvae, and to obtain the 

 fixed stage by keeping them in vessels coated internally with 

 celloidin, but without success. Probably the vessels were too small 

 and shallow, and the water too still to be favourable to their 

 growth. The amphiblastulae remained active in the vessels for 

 twenty-four hours, and it is not likely that this period would be 

 either much longer or shorter under natural conditions, as it will 

 be regulated more by the amount of reserve food material stored 

 within the creature than by small differences in the external con- 

 ditions — so that this gives the young sponge time to drift a con- 

 siderable distance from the parent before it becomes permanently 

 fixed. 



