2 MADRAS FISHERIES BULLETIN [VOL. XVI, 



great care must be exercised to carry this work out economically, 

 for with fisheries few and far between, expenses of inspection 

 mount up rapidly and then wipe out the profits obtained from the 

 fisheries leaving Government and the tax-payer not one pie the 

 better. Governor Van Imhoff long ago sagely queried "Whether 

 there is not more of glitter than of gold about the Pearl Fishery." 

 My own opinion is that the attention of the officer in charge will 

 be most profitably employed if concentrated upon the improvement 

 and development of the always reliable and profitable chank 

 fishery, and that work in regard to the inspection of the pearl banks 

 should be subordinated to and dovetailed with this in order to 

 avoid unproductive expenditure. 



It has been .found unnecessary to amend the original sections 

 except in detail. Two new sections have been added as this 

 appears necessary to make the report self-contained. One of these 

 deals with the anatomy and development of the pearl-oyster ; the 

 other is a concise statement of the conclusions on pearl-formation 

 in our local species to which I have come after twenty years' study 

 of this difficult question. On the other hand the long summary of 

 inspection results from I885 to IQ03 has been omitted as no longer 

 of utility. Similarly a number of the inspection diagrams and 

 plans have been dispensed with. 



The most important occurrence in the history of the Indian 

 pearl banks since 1905 was my discovery in 1913 of a bed of pearl 

 oysters off Tondi, on the west coast of Palk Bay. The fishery that 

 ensued in 1914 was the first ever known in this locality and though 

 it brought little profit to Government, it greatly enlarged our 

 knowledge of the habits of our local species, and cast light upon 

 the hitherto obscure problem of where the beds are which replenish 

 the banks in the Gulf of Mannar after periods when these are 

 absolutely bare of oysters. 



Much attention has also been given since IQ08 to an investiga- 

 tion of the surface drift of the Gulf of Mannar. The results are in 

 the main negative in that they show the drift to be very largely 

 influenced by the monsoon winds. This means that it exhibits no 

 reliable regularity ; instead it varies with the intensity of each 

 individual monsoon and we are no nearer to a knowledge of the 

 factors affecting spat-fall, than that expressed by the statement 

 that the conditions needful to induce a spat-fall depend upon a 

 complex interaction of currents of exceptional character at the 



