The fauna of an inter-tidal mud flat 



179 



substratum to a depth of 10 cm and then pulled out again with its mud core. Usually the consistency 

 of the mud was such that very little of the core was lost. Each of the auger samples was passed separ- 

 ately through a 1 mm sieve and all the organisms retained were preserved in alcohol and sorted later. 



In estimating the total population at any station, the three auger samples were analysed first. Any 

 bivalves exceeding 10 mm length in these samples were discounted, and the numbers of all other 

 organisms then averaged. To this population was added the number of large bivalves only ( > 10 mm) 

 obtained from the im^ quadrat sample at the same station. The whole population was expressed in 

 numbers per m-. 



The shore of North Bay is covered by a soft mud with a little sand, and with shell and gravel some 

 few inches below the surface. There are a few isolated boulders and some patches of weed (mostly 

 Fucus vesiculosus). None of the fourteen sampling stations was sited in weed or near boulders. 

 Even so, the three auger samples at any one station could show considerable differences. Some idea 

 of the great variation which could occur is given for two stations in July 1946, when six auger samples 



Table I 

 Variation in auger samples at two stations: Tr. 4, St. 2 and Tr. 3, St. 4, July 1946. 

 A-F represent actual numbers in 6 samples taken a few feet from each sampling point. 

 The average of 3 samples (A, B and C) has been used for calculating the population 



per m^ 



(A-F) were taken at a few feet distance radiating from the sampling point, instead of the normal 

 three samples taken close together (Table I). The last columns in Table I, showing the average 

 population based on the usual three augers and also an average based on six samples, indicate that 

 the normal three samples probably give a reasonable picture of the average density of a sampling 

 station. In many of the regular series of samples the agreement between the three auger samples was 



remarkably good. 



The efficiency of the sampling by means of a single im" quadrat for the larger bivalves was tested 

 by taking two quadrats at each station instead of the usual one. There was normally reasonable 

 agreement between the pairs of samples (Table II). Occasionally there were large discrepancies (e.g. 

 Tr 1 St 3 and Tr. 3, St. 3), which appear to have been due mainly to a dense but patchy settlement 

 ofMytilus. The bivalve population for July 1946 has been calculated by averaging the two quadrats 



The loss during transport of an auger sample has occasionally made it necessary to calculate the 

 average population at a station on the remaining auger samples. On certain occasions also, at the 

 lowest station of a transect, it was necessary to sample the mud fauna with a small grab, as the area 

 remained covered with water for several days owing to the irregular tides. Generally the whole sampl- 

 ing for the fourteen stations took some three to four weeks. 



