746 DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
having meanwhile assumed an oval form, The cilia do not at first possess the power 
of movement, but in a few hours become actively vibratile and propel the larva 
through the water. As the cilia assume their functional activity the form of the 
body becomes pyriform, the future oral end being the larger. This form is sometimes 
marked in the larva of twenty-four hours (fig. 101), but the difference between the 
oral and aboral extremities is usually less conspicuous than in the specimen figured. 
The swimming movements become very active in the thirty-six-hours’ larva, and 
are very characteristic. The Jarva swims with the aboral end directed forwards, 
revolving at the same time on the longitudinal axis. The larger (oral) end simply 
revolves about its centre while the smaller end describes a circle, so that the larva 
advances by a kind of cork-screw movement. Many of the larvee swim actively 
about, but most of them crowd to the surface, where they arrange themselves in rows 
about the edge with their smaller ends turned upwards and outwards, and the 
swimming movements entirely cease. A very similar habit was observed by Lacaze- 
Duruters in the larve of Astroides. 
By the end of the third day the body becomes elongated (fig. 103), and exceedingly 
contractile and changeable in form. The larva may be at one moment of a worm-like 
elongation and at the next instant contract to a short rounded form as in fig. 105. 
The cilia begin to disappear and the larva swims very sluggishly near the bottom of 
the aquarium. During the fourth day the cilia entirely disappear and the larvz 
sink to the bottom, attaching themselves loosely by means, apparently, of a mucous- 
like secretion. The larval life is now ended and the tentacles and spicules soon make 
their appearance (see § 11). 
Leptogorgia agrees in the main with Renilla, but the development takes place more 
slowly. The embryo, after passing through the period of distortion, becomes of a 
regular oval form and acquires a uniform coating of cilia. The aboral end soon 
becomes slightly smaller and the larva swims with the same peculiar cork-screw move- 
ment observed in Renilla. The larvee have the same habit of arranging themselves in 
rows at the surface of the water. On the fourth day the larve are much elongated 
(fig. 112), and possess the power of active contraction, The larval life is not ended 
until about the sixth or seventh day, when the cilia disappear, the larva sinks to the 
bottom and once more assumes a short rounded form (fig. 113), and the eight septa 
become faintly visible about the eighth day. Some of the larve attach themselves 
firmly by the aboral end, but others remain free as long as they were kept under 
observation (seven weeks). In one case two larve, originally quite distinct, became 
attached to each other near their oral ends (fig. 114). The union became very 
complete in a day or two, and no line of division between them could be made out. 
The larve were kept for a fortnight, but underwent very little change, and finally 
died. I believe their union was due simply to accidental adhesion, and has no signi- 
ficauce bearing upon the formation of the colony. KowALeysky observed in 
Alcyonium that numbers of the larvee fused together in a similar manner, but their 
