46 D. A. HANCOCK AND A. C. SIMPSON 



from the greater vulnerability of young to physical conditions, selective 

 feeding of predators may have its effect. This may result from actual prefer- 

 ence, e.g. of the American tingle for young oysters, or the physical inabihty 

 of certain predators like starfish to eat bivalves larger than a certain size. For 

 example, Thorson (1955) found that young lamellibranchs were able to 

 survive in areas containing vast numbers of predatory brittle-stars. This was 

 because the brittle-stars did not feed during their breeding period, during 

 which the lamellibranchs were able to grow too large to be ingested. 



The causes and extent of mortahty can often be estimated from the shells 

 of dead molluscs which remain on the bottom. The numbers of live and dead 

 oysters in a dredge may be compared, provided the time interval since death 

 of the shell or 'clock' is known. Some predators leave characteristic marks, 

 such as the holes drilled by the American tingle in oysters (Hancock, 1954), 

 or the edges of oyster shells broken by the attacks of starfish (Hancock, 

 I955)» or shells of cockles opened by oystercatchers (Drinnan, 1957). The 

 dead shells of oysters may contain signs of disease, like shell disease or boring 

 worm, which may have contributed to their death. The shells of dead whelks 

 often become inhabited by hermit crabs, which are also caught in baited pots 

 with whelks. Many hermit crabs with painted shells were recaptured in a 

 whelk marking experiment, and with some knowledge of the habits of the 

 hermit crabs, their numbers might be used as a guide to post-marking and 

 subsequent natural mortality. 



SUMMARY 



The important parameters required for making yield estimates in exploited 

 invertebrate populations have been discussed. Certain features appear to be 

 characteristic of shellfish populations and of the methods used for estimating 

 these parameters. These are briefly : 



(i) The ease of sampling of exposed sedentary populations, which allows 

 separate estimates to be made of total density, magnitude of recruitment, 

 fishing mortahty and natural mortality. 



(2) Special problems associated with the use of baited traps. 



(3) Growth estimates are made easy in a number of molluscs by the presence 

 of clear growth rings, but are extremely complicated in Crustacea as a result 

 of growth by moulting. 



(4) Recruitment is very irregular among many molluscs, and mortalities 

 may be catastrophic at all stages of the life history. 



(5) Yield assessments require consideration of seasonal changes within the 

 exoskeleton. ^ 



