a1 Wellcome Marine Laboratory and The Department of Pure and Applied Zoology, University of Leeds, Leeds, LS2 9JT
The numerical relationships between a population of Lepeophtheirus pectoralis and its host population (plaice) have been described. There was no significant difference between infection levels in male and female plaice. In most of the monthly samples (75%) there was no significant relationship between host length and either the incidence or intensity of infection. The samples in which significant positive correlations occurred were usually taken within the period November to April. It is probable that the immigration of smaller fish, with lower levels of infection, from shallower water produces this effect.
The parasite population was found to have an overdispersed distribution (variance > mean) within its habitat, the host population. The frequency distributions of numbers of parasites per fish were calculated and the negative binomial distribution was fitted to these. The fit was good in 22 of the 24 samples and the value of k, the exponent, was found to be directly related to the mean density of the parasite population. Iwao's (1970) method of dispersion analysis indicated that the observed negative binomial was generated by a model of randomly distributed clumps or colonies of individuals. A possible mechanism of producing this pattern was proposed based on evidence from studies of mortality in the parasite population (which indicated that all-or-none type mortality operates on the clumps at certain periods during the year), the infection process (which demonstrated that a single wave of infection produces an overdispersed spatial distribution) and the examination of parasites from individual fish (which displayed a clumping of parasites at certain stages rather than an even spread over the range of possible stages). The mechanism proposed is that the adaptations for host location of the infective copepodid stage tend to produce aggregations of these larvae in favourable areas of a heterogeneous habitat. Fish moving randomly into these areas will tend to be infected by more than one larva (a clump of larvae) at any one time when available for infection. The fish can be regarded as equivalent to random samples of the free larval population and the aggregated distribution of the parasite population on its hosts is therefore directly related to the aggregated distribution of the free-swimming infective larvae.
(Received June 04 1974)
p1 Department of Zoology, British Musieum, (N.H.), Cromwell Road, London SW7 5BD.