FISHERY BULLETIN: VOL. 81, NO. 2 



using numerous parasite groups, the parasite taxa 

 that do not parallel the host phylogeny are likely to 

 become apparent. 

 Another parasitological theory we have tested is 

 "Szidat's rule," which suggests that primitive (gen- 

 eralized) parasites are found on primitive hosts and 

 that advanced (specialized) parasites are found on 

 advanced hosts. We provide an example supporting 

 this concept when we consider the scombrid host 

 preferences of the copepod genus Unicolax on scom- 

 brid hosts (p. 254). 



SAMPLING ADEQUACY AND 

 HOST SIZE 



Before considering host specificity, it is necessary 

 to know whether enough hosts were examined to pro- 

 vide samples of all species of the usual parasite 

 fauna. Individual collections of copepods from each 

 scombrid species were recorded on cards sequential- 

 ly, enabling us to consider the question: "How many 

 specimens of a host species should be examined 

 before all parasitic copepod species are likely to 

 have been collected?" Examples are given in Ta- 

 ble 1. 



Table 1 . — Number of specimens that had to be examined in order to 

 find all known copepod species. 



Of the six species presented in Table 1, the two spe- 

 cies of Scomberomorus endemic to the eastern Pa- 

 cific (S. sierra and S. concolor) required a relatively 

 small number of individuals to be examined (2-12 

 specimens), until all parasitic copepods were collect- 

 ed. Wider ranging species (S. commerson, Sarda sar- 

 da, Euthynnus affinis, and Auxis spp.) required 



FIGURE 1. — Examples of copepods parasitic on scombrids: a) Uni- 

 colax anonymous, female; b) Holobomolochus asperatus, female; c) 

 Shiinoa inauris, female and males; d)Caligus bonito, female; e) Ely- 

 trophora brachyptera, female; f) Gloiopotes hygomianus , female; g) 

 Tuxophorus cybii, female; h) Pseudocycnus appendiculatus , female; i) 

 Pseudocycnoid.es armatus, female; j) Brachiella thynni, female and 

 dwarf male attached. 



examination of a greater number of specimens (35- 

 60) before we collected all of their copepod species. 

 The two endemic species have fewer species of 

 parasitic copepods than the nonendemic species. 

 Other scombrids with restricted distributions 

 (Scomberomorus multiradiatus , S. sinensis, and S. 

 munroi) also have fewer parasitic copepod species 

 than related species with wider distributions. 



When collecting parasitic copepods from hosts with 

 wide distributions, specimens must be examined 

 from throughout the range. We found that the num- 

 ber of parasite species is usually less at the periphery 

 of the host's range, so that conclusions relative to to- 

 tal parasite fauna for a species cannot be based on 

 geographically limited collections. 



We also examined the relationship between host 

 size and infestation density in order to determine its 

 importance in sampling adequacy. It is generally ac- 

 cepted that larger individuals of host species usually 

 support a greater parasite fauna, both in number of 

 species and individuals. Although little work has 

 been done on the ectoparasite fauna in relation to 

 host size (age), Dogiel et al. (1961:9) noted an in- 

 crease in the numbers of Ergasilus sp. on the gills of 

 Esox lucius on larger fish. Cressey and Collette 

 (1970) found that specialized copepods (those pos- 

 sessing holdfasts or that are very host specific) are 

 found mainly on larger needlefish, while generalized 

 copepods (less host specific and not highly modified) 

 are found most often on smaller needlefish indi- 

 viduals. 



In the present study, copepods of the families 

 Pseudocycnidae, Bomolochidae, and Shiinoidae 

 parasitic on three species of Scomberomorus were 

 considered (Table 2). We chose these copepod spe- 

 cies for the study because they remain attached in 

 preserved specimens. Pseudocycnids (Fig. lh, i) are 

 firmly attached to gill filaments; bomolochids (Fig. 



Table 2. 



-Infestation densities of Scomberomorus commerson, S. maculatus, and S. brasiliensis 

 for three copepod groups, Pseudocycnoides, Bomolochidae, and Shiinoa. 



228 



