The production of Liza ramada (Risso) in Lake Mariut. Egypt 253 



associated with unexpectedly low catches. Inspection of the graph gives the impression 

 that catch rises roughly in proportion to fry introduced up to about 25 million, but 

 that two higher values than this correspond to lower catches, perhaps due to high 

 mortality from more acute competition among the more numerous fry. 



A correspondence between the catch and the number of fry introduced is not 

 difficult to imagine, and it seems reasonable to suppose that it could happen in two 

 ways. First of all it is to be noted that a substantial yield of catchable fish, 47 tons per 

 4-7 million fry, occurred in the year of first introduction, and it is reported that a 

 number of these left the lake to spawn in the first year. The introduction of fry in 

 any one year may therefore be expected to have a substantial effect on the yield. 

 Secondly, as a fair proportion of the fish spawn in their first year, a successful year- 

 group may be indicated by a good catch and the origin of a successful spawning. This 

 would mean there might be a correspondence between the catch in one year and the 

 number of fry that appeared in the succeeding year. 



In fact when the catch and contemporary fry were correlated, a low but positive 

 correlation of little significance was obtained. It is not without interest, in view of the 

 preceding remarks, to report that for the relation between the catch and the fry 

 introduced a year later the correlation was somewhat higher. 



From age, length, and weight samples given in Table II, and in the Egyptian Fisheries 

 Reports for 1920, 1928-31, and 1933, estimates of the numbers of O group fish caught 

 in 1920, 1928-31 and 1933 have been devised and shown in Table I. The numbers 

 have been obtained by dividing the estimated mean individual weight into the weight 

 of the Tobar landings for the periods August-December of the appropriate years. 

 These estimates may be compared with the corresponding fry introductions. 



It will be seen that in 1920 nearly a million fish of commercial size were taken out 

 of the lake within nine months of introducing about 4-7 millions. This is to say that 

 it needed rather less than five fry to produce a marketable O group fish. By 1928 it 

 took 50 fry to do this, and the O group caught had fallen to about 400,000— a fall 

 which continued steadily until 1933. The number of fry necessary to produce one 

 commercial O group fish increased until in 1930 it was nearly 800, but thereafter it 

 went back to 269 in 1931 and 132 in 1933. These observations suggest that the 

 conditions for survival must have deteriorated throughout the period. Against this 

 it must be remarked that the estimated proportions of O group fish in the August- 

 December catch, where they may form part of the spawning shoals, fell from 42% 

 to 17% between 1928 and 1935, and it may be that more O group stayed on in the 

 lake to become 1 group in this period. Nevertheless, from 1929 to 1931, when the 

 percentage of O group was steady, it appears that the O group numbers continued to 

 fall, and survival conditions were still getting worse. 



Catch, Effort and Lake-level 



The catch (Fig. 1) may be broken up into five periods, three of adjustment and 

 change and two in which some stability is indicated. These are. first a period o( 

 increase from 1920 to 1923, during which the fry introductions built up the popula- 

 tion followed from 1923 to 1926 by a period of stability during which the catch 

 fluctuated around a level of rather under a 100 tons. Then in 1927 there was a rise 

 immediately succeeded by a sharp fall through 1928 and 1929 to 1930. From 1930 to 



