THE LIFE-HISTORY AND HABITS OF THE PEARL OYSTER, 127 



on the Algse. So that we have seen, either in our experimental tanks or in collections 

 from the sea, every stage from the egg to the adult pearl oyster. Mr. Hornell has 

 made a careful study of the young stages as observed by him at Galle in May, 1902, 

 but as we hope to get still further details, the full account of the embryology will be 

 given later on in this report. It will suffice to state now that the ovum when extruded 

 is pyriform and floats, after fertilization it becomes spherical. The segmentation is 

 complete but unequal, and within 2 hours results in an embryo of one large 

 macromere and six micromeres. After 4 hours, segmentation was completed and 

 the embryos were swimming freely by means of cilia, while about 20 hours after 

 fertilization they were in the trochosphere stage, with a well-marked pre-equatorial 

 band of long cilia, an apical pre-oral tuft and a patch or circlet of cilia at the opposite 

 pole. As the body elongates, the equatorial band moves further and further forwards 

 and becomes a pre-oral circlet, most strongly marked along the anterior margin which 

 becomes the velum. Towards the end of the second day the larva is a veliger, and 

 the shell has commenced to form posteriorly. On the third day the velum was 

 considerably reduced in size and the bipartite shell increased, and this brought the 

 larva to a stage corresponding with the forms we caught in our tow-nets, which 

 bridged over the gap between the latest stage reared from the egg in the tanks and 

 the earliest attached stage or " spat." 



From a consideration of the sizes of the free-swimming larvae and the youngest spat, 

 and knowing the age at which the veliger obtains a shell, we are of opinion that 

 attachment may be made within five days of fertilization. At the same time, from the 

 large size of some of the free-swimming larvse, it is probable that under certain 

 circumstances, such as the absence of suitable areas for attachment, the period of 

 free-swimming existence may be considerably extended. 



One of the earliest free-swimming stages was taken in the tow-net on February 2nd, 

 1903, on Modragam Paar. It shows some slight advance upon the oldest stage reared 

 in the Galle Laboratory, as rudiments of three branchial filaments and of the otocysts 

 are present. Otherwise, in the form of the shell, the large size of the velum and the 

 sub-central position of the digestive gland the two agree. The digestive >land is 

 yellowish and granular, the cells being filled with large oil-globules. 



In the same haul numbers of an older stage (fig. 41,1.), nearly twice the diameter of 

 the last, were also present. In these the shell had developed prominent umbones 

 placed equidistant from the two ends. A few more branchial filaments had appeared 

 and the velum was relatively smaller. The digestive gland is now conspicuously two- 

 lobed, and by great increase in size has come to occupy the interior of the umbones 

 dorsally as well as extending ventrally beyond the plane of the otocysts. The 

 rudiment of the adductor muscle at the anterior end was also visible. 



A fortnight later (16th February), while on the Cheval Paar, in plankton taken 

 between 8 a.m. and noon, we obtained a great multitude of shelled larva?, many of 

 which were in a later stage. The shell-valves showed for the first time a slio-ht 



