responsible for the greater number of larvae setting 

 on the upper siu-faces. 



A lack of consistency in observations of various 

 investigators in different environments indicates 

 that it is impossible to ascertain the effects of a 

 single factor of the environment while testing 

 under complex and variable natural conditions. 

 Real progress in the study of the reaction of 

 oyster larvae may be achieved if further observa- 

 tions are made under controlled conditions. 



Larvae of C. virginica that are grown artificially 

 in culture jars and not disturbed by stirrmg or 

 aeration are more or less uniformly distributed. 

 Eyed larvae frequently congregate on the sm-face, 

 swimming with their vela uppermost and touching 

 one onother with the tips of the cilia. They 

 form groups or "rafts" visible to the naked eye. 

 Some of them close their valves, fall rapidly to the 

 bottom, and after a short time resume swhirming. 

 Falling to the bottom should not be confused with 

 negative geotaxis, which has not been demon- 

 strated for oyster larva. In the laboratory, larvae 

 often attach themselves to the sides of plastic or 

 glass containers and apparently do not discrmii- 

 nate between light and dark surfaces. 



TO THE PROPERTIES OF SURFACE 



Oyster larvae attach themselves to many kinds 

 of hard and semihard surfaces. They are found 

 on rocks, gravel, cement, wood, shells of other 

 mollusks, on stems and leaves of marsh grass, and 

 on a great variety of miscellaneous objects such 

 as tin cans, rubber boots and tires, glass, tar paper, 

 and pieces of plastic that may be accidentally 

 thrown on the bottom or deliberately used as spat 

 collectors. There is no evidence that the larvae 

 are selective in finding a suitable place to set, 

 provided the surface is not covered with a slimy 

 film, detritus, or soft mud. Under natural con- 

 ditions they are never found on shifting sand or 

 on a bottom covered with loose sediment. Success 

 of setting always depends primarily on the avail- 

 ability of clean surfaces rather than on other 

 factors. Shells covered with oil and greasy sub- 

 stances in polluted areas are not suitable for the 

 attachment of larvae. Cole and Knight-Jones 

 (1939) found that it is difficult to induce 0. edulis 

 to set on smooth glass, but the larvae of C. 

 virginica raised in the laboratory readily attach 

 to polished glass. In fact live preparations of 

 spat may be obtamed for microscopic examination 



of small oysters by suspending glass slides in a 

 tank with fuUy grown larvae. 



GREGARIOUSNESS 



An interesting gregarious tendency has been 

 observed by Cole and Knight-Jones (1949) among 

 the larvae of 0. edulis. During experunents m 

 large rearing tanks they found that larvae set 

 more readily on shells akeady bearmg 50 to 100 

 spat than on shells bearing fewer spat. They 

 suogest that a substance secreted into the siir- 

 rounding water by the spat, and possibly by the 

 fully developed larvae, encourages the setting. 

 No attempts were made to isolate the substance 

 and test its effect. The authors make another 

 observation which may throw some doubt on the 

 validity of their interpretation. They state (p. 36) 

 that "Larvae set more readUy on shells which 

 had remained uncleaned in the tanks for 2 or 

 more weeks, and which bore a \'isible film of 

 bacteria or diatoms, than on similar shells which 

 were cleaned daily." In their study of gregarious- 

 ness they placed shells of uniform size and shape 

 in pairs in a tank containing fully developed 

 larvae, and the number of spat attached to them 

 was counted daily. One shell of the pair was con- 

 sidered a control and was cleaned every day, and 

 the other (experimental) remained uncleaned. By 

 tlie end of the setting period the total spat settled 

 on the experimental shells significantly exceeded 

 the total spat settled on the controls by a ratio of 

 2.5 to 1. The figures suggest that the observed 

 differences may be due to the attraction of larvae 

 by those which had already settled on the shell, 

 but the conclusion cannot be accepted without 

 further verification. The possibility is not ex- 

 cluded that some other unknown factor, such as 

 tlie position of the controls in relation to the ex- 

 perimentals, affected the results or that handling 

 and removal of spat from the control shells caused 

 changes to the surface which made them less 

 attractive to the larvae. It would be profitable 

 to conduct a series of tests designed to eliminate 

 bias by placing experimental and control shells at 

 random and making a statistical analysis of the 

 significance of the differences. 



Yonge (1960) expresses no doubt "that larvae 

 (of O. edulis) settle more readily on surfaces to 

 which others are already attached," and points 

 out that this tendency aids in reproductive effi- 

 ciency and is, therefore, a major benefit to attached 

 animals. In view of the fact that a single oyster 



L.'iRV.'iL DE\ELOPMEXT AND METAMORPHOSIS 



373 



