OF PSEUDOCOCCIDAE & ERIOCOCCIDAE 31 



DISCUSSION 



This study of the male representatives of Pseudococcidae and Eriococcidae has 

 revealed many interesting facts regarding their taxonomic status. Moreover, the 

 material of Pseudoccidae, being fairly representative (20 spp. were considered), 

 allowed for certain suggestions as to the classification within the family. The 

 material of Eriococcidae, although inadequate for similar suggestions (7 spp. only), 

 was sufficient for a general characterization of the males of the family, and for 

 suggestions regarding the various degrees of morphological relationships between 

 the studied species. 



These conclusions were assessed by taking into consideration as many characters 

 as possible, as listed in table I. The table includes all the species here studied and 

 3 others adequately described for the purpose of comparison by Giliomee (1961). 

 The large number of species studied by Beardsley (i960, 62, 63, 64 and 65) could 

 not be included in this table since his descriptions contain comparatively few of the 

 characters here employed. The assessment of the relationships between the studied 

 species was carried out by using the quantitative evaluation, method adopted by 

 Ghauri (1962) and followed by Giliomee (1967). In this method all the characters 

 were regarded as being of equal importance, and those shared by any two species 

 were counted, and the calculated numbers (shown in table II in the form of a 

 matrix) were taken as indices of the degree of affinity. This method was found 

 most practical in the circumstances, although not the most satisfactory from the 

 statistical point of view. The methods recently discussed by Sokal and Sneath 

 (1963), in which a correlation coefficient is obtained for each pair of species, is more 

 accurate, but involves the use of an electronic computer for calculation and the 

 development of a rather elaborate computer programme.* 



The discussion, therefore, will include the following aspects : 



(a) The taxonomic status of Pseudococcidae and Eriococcidae. 



(b) The taxonomic significance of the systematically important characters, and 

 the levels at which they could be used. 



(c) The classification and interrelationships within each of these groups. 



(d) The relationships of the Pseudococcidae and Eriococcidae with other 

 families of Coccoidea. 



(A) Status of Pseudococcidae and Eriococcidae 



On examining table I, it was immediately apparent that the pseudococcid and 

 the eriococcid males exhibit a number of structural differences, comparable with 

 those separating other families (e.g. Pseudoccidae and Coccidae), thus justifying the 

 recognition of the two groups as distinct families. This conclusion is in accordance 

 with Ferris's (1937) concepts of classification, which were also adopted by Hoy 

 (1962-63). 



* This programme has only recently been developed in the Department of Zoology and Applied 

 Entomology by Dr. G. Murdie, taking the advantage of the computing facilities available in the Imperial 

 College. It was applied too late to include in the main part of the thesis, but the results of the two 

 methods are discussed under Analysis. 



