106 K. Takahaslii 



projection which is almost as long as the 2ncl antennal joint, with 2 capitate 

 liairs at the apex; the 3rc] slightly inibi'icated on the distal portion, without senso- 

 ria; the 4th, 5th and 6th imhricated; the relative length of joints about as follows: 

 111-40, IV-27, V-30, VI-Tf) (20+55). Eostrum reaching beyond the middle 

 coxae? Capitate hairs on tlie abdomen as long as the 2nd antennal joint, but those 

 on the hind part of the abdomen iniuii longer. Cornicles long, as long as the 6th 

 ■antennal joint (base and spur), moderately dilated on the distal lialf and at the base, 

 broadest at tlie base, not curved, scarcely imbricated, not reticulated. Cauda stout, 

 much sliorter than the cornicle. Legs long and slender, provided with many rather 

 long scarcely knolibed hairs; liind tarsi a little shorter tlian the base of the last 

 antennal joint. 



Lengtli of body-about 1.0 mm. Antenna-about 1.0 mm. 

 Cornicle-about 0.3 mm. 



Host.-unknown in Ja])an. 



Uistribution.-Formosa, Japan. 



A wingless viviparous, female was collected by tlie autlior on June 4, 1923, at 

 Kyoto, Japan. Hitherto unrecorded from Japan. Tlio Japanese specimen differs 

 somewhat from the Formosan specimens, but seems to belong to this species. 

 Myzus (?) polygoni (v. d. Goot). 

 (PI. IX, A, fig. 9 and B, figs. 1-2). 



PI lorodon polygoni, van der Goot, C ntrib. Fauna Indcs Xeerland., 1, 3, j>. 

 44(1918). 



'Rost. -Polygonum sp. 



Distribution .-Japan: Tokyo, Kyoto. Java. 



INIany wingless vivip.irous females were collected in June 1923. Hitherto 

 unrecorded from Japan. 



The Japanese specimens agree almost exactly with the original description, 

 diSering from it, however, in the following characters: 



(1) Yellow, cornicles yellow. 



(2) The last antennal joint sliorter; the relative Icngtli of antennal joints 

 about as follows: III-85, IV-63, V-57, V- 128 (23-|-105). 



(3) Rostrum shorter, reaching the middle coxae. 



This aphis is not a typical 3Tyzufi and seems to represent a new genus, but tlie 

 author hesitates to propose a new generic name for it, since the winged form has 

 never been collected. 



