i8o FAUNA HAU'AIIENSIS 



what I have observed about our Hawaiian forms. As the results differ in the case of 

 the three groups, I will take them separately. 



Group Anchomenides. A fair standard of comparison as to the size of the 

 wings can be found in this group by comparing the length of the wing with that 

 of the elytron. Most of the Hawaiian winged genera of this group have the wings 

 about the usual size, the length of the wing being about i^ that of the elytron. In 

 Baryneiis sharpi the measurement gives wing 15 mm., elytron 9^ mm. Although no 

 data have, so far as I am aware, been published previously on the subject, this is I 

 believe about the same relation as is usual outside the islands ; I find that in our 

 well-known European Anchovienus paniiupjinctatus, the lengths are wing 6| mm., 

 elytron \\ mm. The Hawaiian genera Colpocaccus, Mysticomeims, and Colpodiscus, 

 as well as Bnryneus, have similar relative measurements. In the genus Chalcomemis 

 the dimension of the wing is distinctly reduced, it being in C. molokaiensis 7 mm. 

 to 5i for the elytron. In the aberrant C. costatus (from Kauai) the reduction of 

 the wing is strongly marked, as is evident from inspection of PI. VI. fig. 22. 

 Measurement here gives wing 5f , elytra 4| mm. The wing is in this case however 

 perfectly well developed, all the nervures being present and strong. There is no 

 connecting link between this and the numerous forms of vestigial wings. The.se 

 vary much in size according to the species, as may be seen by a glance at the 

 plate ; the transverse folding is completely absent, and the nervuration is very 

 incomplete ; the apical portion of the wing has in fact completely disappeared. The 

 vestiges differ to some e.xtent in shape, and a little in nervuration ; the stigma, or 

 large chitinous spot on the costa, can frequently be detected just at the tip of the 

 vestige, and differs a little in position. The vestige is large in Barypristus in- 

 cendiarms and also in Aptcromestis vmatlatus : in the former measurement gives 

 about, wing 6\, elytron \o\\ and in Aptcromcsus, wing 2, elytron 4. The vestiges 

 are extremely reduced in the most remarkable of the Hawaiian Anchomenides, being in 

 Dcropristiis about \ mm. long, while the elytron is 5f mm. 



In any single species of Anchomenid the vestiges vary but little so far as I have 

 observed. In testing this I have had assistance from Miss Alice Embleton. Upwards 

 of fifty e.\amples of one species have been examined, the method being as follows. 

 The vestiges not being visible in the ordinary condition of the insect with closed elytra, 

 38 examples of these were selected ; the chosen being those that differed most in size 

 and shape : one elytron and one vestigial wing (those of the right-hand side) were 

 taken off each specimen and fastened with gum on ruled millimetre paper, specimen 

 and appendages side by side. In addition to the examples measured and tabulated 

 by Miss Embleton, I myself dealt with fifteen or sixteen others. In all upwards of fifty 

 specimens of Barypristus rupicola were dealt with. It is scarcely necessary to give the 

 measurements of all the individuals, it is sufficient to say that the elytra were found 

 to vary in length about 27 mm., the longest being i r2, the shortest 8'5. The vestiges 



