78 MORPHOLOGY AND TAXONOMY OF ADULT MALES 



From the same tables it appears that Giliomee was justified in assuming that the 

 Coccidae are more specialized than Pseudococcidae. They have 5 specialized and 

 6 primitive features against 4 specialized and also 6 primitive in the latter, and 

 comparison with Diaspididae shows that they share a total number of 7 characters 

 with Coccidae but only 3 with Pseudococcidae. 



Table VIA also indicates that Coccidae are closer to Eriococcidae (sharing 5 

 characters) than they are to Pseudococcidae (sharing only 3 characters). In the 

 same way it also appears that the Eriococcidae are more closely related to the 

 Pseudococcidae (sharing 9 characters) than they are to the Coccidae (sharing only 

 5 characters). 



Table VIB is also derived from table VI and prepared in the form of a matrix. 

 Here the relationships between the groups are expressed in fractions obtained by 

 dividing the number of shared conditions by the total number of characters con- 

 sidered. Those characters occurring in either specialized or primitive conditions 

 were counted as 1 unit, and if both conditions occurred in any of the compared 



Table VIB 





Steingelia 



Phenaco- 

 leachiidae 



Pseudo- 

 coccidae 



Erio- 

 coccidae 



Coccidae 



Diaspi- 

 didae 



Margarodidae 



(073) 



(073) 



(0-62) 



(0-58) 



(050) 



(0-42) 



Steingelia 





(062) 



(050) 



(062) 



(077) 



(0-46) 



Phenacoleachiidae 







(o-88) 



(077) 



(0-46) 



(0-46) 



Pseudococcidae 









(o-8i) 



(0-42) 



(o-35) 



Eriococcidae 











(0-46) 



(0-54) 



Coccidae 













(0-62) 



Diaspididae 



groups, they were both given a score of \ unit each, e.g. Margarodidae and Steingelia 

 share 9 characters and are separated by 3, but in the former there is one character 

 (No. 7) occurring in both conditions ; the numbers thus corrected will be 9J shared 

 and 3! separating. The number of the shared characters (9!) was divided by the 

 total number of characters (13) and the result (073) was entered in the table as a 

 measure of similarity. The values obtained vary from 0-1 indicating no or complete 

 similarity respectively, i.e. the higher the values are, the closer the relationship is. 



From this table it is obvious that Steingelia and Phenacoleachiidae are similarly 

 related to Margarodidae, and are closer to it than they are to each other. It is also 

 apparent that Phenacoleachiidae are more closely related to Pseudococcidae than 

 they are to Margarodidae, which would support the views of Beardsley. It also 

 appears that Steingelia is closer to Coccidae than to the Margarodidae. 



Giliomee (1964) discussed the taxonomic status of the family Kermococcidae, which 

 Ferris (1937) placed in the family Eriococcidae, but which Balachowsky (1942) 

 included as a subfamily in one family (Eriococcidae) together with Pseudococcinae 



