1969] 
Matthews — Scarabaeine Beetles 
123 
Genus Canthonella Chapin, 1930 
This genus was recently redescribed in detail and its affinities 
discussed by Halffter and Martinez (1967). They showed that the 
genus Ipselissus Olsoufieff, of Brazil and Northern Argentina, is 
extremely closely related to Canthonella, sharing even the very un- 
usual form of the male genital capsule. Canthonella , therefore, is 
not as isolated taxonomically as the present author believed and is 
not of direct Holarctic origin in the Hemisphere. Canthonella and 
Ipselissus may be imagined to be now-localized relicts of a once- 
widespread Neotropical genus. 
Less easy to explain is the similarity of both these genera to the 
genera Sauvagesinella Paulian of Southwestern Australia and 
Nesovinsonia Martinez and Pereira of Mauritius, described by 
Halffter and Martinez (op. cit.) and previously hinted at by the 
present author (Matthews, 1966:76). The genus Sauvagesinella 
needs to be redescribed in detail, a task which the present author will 
accomplish in his current revision of the Australian Scarabaeinae, 
during which affinities with Neotropical groups will also be examined. 
Canthonella pygmaea (Harold, 1869). 
The lost type of Canthon pygmaeus Harold was found by the 
author quite by accident in an unmarked box in the Paris Museum. 
Thi s box contained numerous American canthonine specimens, in- 
cluding a number of unique types, removed from the Rene Oberthiir 
collection, probably by R. Paulian. The other types found will be 
reported on by G. Halffter and A. Martinez in their current revision 
of the American Canthonina. 
The type of this name belongs in fact to the Cuban species which 
has gone under the name previously (Zayas and Matthews, 1966; 
Matthews, 1966) and therefore presents no surprises. It bears the 
label, in Harold’s own handwriting: “ pygmaeus typ. Harold”. It 
has now been placed in the Collection Generale near the types of 
histeroides and gundlachi (now in Canthochilum ) . 
Vulcano and Pereira (1966:120-122) cite Canthonella pygmaea 
as occurring in Puerto Rico and Hispaniola, as well as Cuba. The 
Hispaniolan specimen on which they base this record has been ex- 
amined by the present author and found to be Canthonella constans 
Matthews (mentioned below). The two supposed Puerto Rican 
specimens have not been seen by the author, but a total of 184 
specimens of Canthonella, collected from all over the island of Puerto 
Rico, has been examined and all are C. parva Chapin. If the two 
specimens cited by Vulcano and Pereira are indeed pygmaea, then it 
