BIOGEOGRAPHICAL BOUNDARIES 



21' 



• — Ceratium fusus 



O — Ceratium macroceros 



A —Calanus finmarchicus 



I I I I I I 

 1948 A9 50 51 52 53 54 55 56 57 



Fig. 8. Annual fluctuations in abundance of two dinoflagellates and a copepod 

 in the southern North Sea, taken with the Continuous Plankton Recorder. The 

 average numbers per sample are expressed on a common measure scale about the 

 mean number for the whole period of 10 years (G. A. Robinson and J. M. Cole- 

 brook, unpublished). 



relatively small number of patterns of abundance fluctuations of 

 this kind. From an analysis of a matrix of correlation coefficients 

 it should be possible to establish which groups of organisms fluc- 

 tuated in similar or in opposed ways. This is an extension of the 

 technique of determining associations between species from a 

 matrix of coincidence of pairs of species in a group of samples. 



Figure 9 shows another application of the study of these annual 

 fluctuations in abundance. I should explain that one of the most 

 interesting events in the herring populations of the southern North 

 Sea was the increase in length which occurred in 1950. Gushing and 

 Burd (1957) showed that this was contemporary with an increase in 

 the abundance of Cala?iiis finmarchicus, an important food of the 

 herring. The lengths of the fish remained high, and the Calanus 

 continued to be abundant until 1956 when a drop in the length of 

 the herrings was coincident with a fall in the numbers of Calanus. 

 This is illustrated here, by using the mean length of three-year 



