162 DuNN—NOTES ON CHINESE LABIATAE. 
fortiter to—nervius, aliter similis. Corolla alba, calyx vix longior, 
apice breviter bilabiata, antice barbata. Nuculae 2.2 mm. diam. 
globosae, albae, alte reticulatae. 
JAPAN. Cult. under name of Ygoma (fide Savatier) ; Nagasaki 
Maximowicz. 
CHINA. Curt1; Peking, Bretschneider 577; Paushan (Peking), 
‘“Sutzu, bird seed and oil plant,’ October 1, 1886, 
Carles, 140. 
FoKIEN. Foochow, Carles, 733. 
Yunnan. ‘S.W. China, Shu ma, seeds used in cooking,” 
Bourne, 2; Salwen and Irrawady basins, lat. 25° to 27° N. at 
4~7000 ft. and in Upper Burmah, grown by natives (Shans and 
Lissoos) for an oil which is expressed from the seeds and used for 
cooking, Forrest, 871 ; Yunnan-sen, Maire 25, 734, 2614. 
P. Cavaleriet, Léveillé in Fedde, Repert. Nov. Sp. viii (1910), 
425. Typein Le Mans Herb.=P. nankinensis, Hemsl. 
Mosla Argyt, Léveillé, I.c-ix, 247. Type in Le Mans Herb.= 
M. lanceolata, Benth. 
M. Fordit, Maxim. in Mél. Biol. xii, 525. The type in the Kew 
Herb.=M. chinensis, Maxim. 
The latter was described from specimens which Augustino- 
wicz gathered near Foochow, and the type specimens are in 
the St. Petersburg herbarium. Maximowicz afterwards received 
specimens of an allied form gathered by Ford in the Lo-fou-shan 
and described them as M. Fordti, relying on the silkiness of the 
floral bracts and their abruptly acuminate tips to distinguish them 
from the Foochow plant. The series of specimens in the Edin- 
burgh herbarium together with those at Kew have now been 
compared with Maximowicz’s type specimens, accompanied by 
beautiful drawings, courteously lent to Kew for the purpose, and 
it is clear that specific distinction should no longer be maintained 
between the two forms. Carles 725 from Foochow is about in- 
termediate between the two types, having the silkiness of M. 
Fordit, with the bracts of M. chinensis. 
M. punctata, Maxim.=dianthera, Maxim. 
The writer does not hesitate to acknowledge, being in this 
respect in company with Hemsley, that he is unable to sort the 
material of Mos/a entirely according to the proposals of Maxi- 
mowicz. With the exception of the size of the bracts, which 
almost cover the flowers in M. chinensis, no constant characters 
have been found besides those of the seeds. The two forms 
here referred to (M. punctata and M. dianthera) have seeds of 
the same character, loosely and shallowly reticulate, and in the 
large series of specimens of them now collected at Kew from 
