126 
tion of the species. In the Philippine Islands thirteen distinct 
and well-marked species have been described and at least six 
+: 
has been collected in Mindanao, Leyte, Samar and Biliran. But 
these islands are not far apart and, moreover, the w a distribu- 
tion may be explainable on the grounds of the economic im- 
portance of the seeds. Whether S. multiflora is confined to the 
Island of Luzon or whether it also occurs in some other islands 
is still doubtful; and until better specimens. nite fruits can be 
aperie of the multiflora- hike plants from Mindanao, Mindoro, 
, the point must remain unsettled.* A re- examination of the 
specimens however, suggests that the species is probably endemic 
n Luzo At present we know the following endemic species 
in these idands 
Luzon.—S. multiflora, S. hed eee! S. Merrillii, and five 
ssipantane but specail distinct spec 
Mindanao.—S,. [gnatii, S. dubia, § ee S. similis. 
Paragua.—sS. ovata, S. oleifolia, S. impressinervis, and one 
imperfect but finite distinct plant. 
Leyte.—.S. Iqnatii, S. Wenzelit. 
Samar.—S. Ignatii, S. tesseroidea. 
Biliran.—S. [gnatii. 
Panay.—S. panayensis. 
Ticao, Cution and Mindoro are each represented by distinct 
but insufficiently known specimens in our herbaria 
The Philippine species belong to all four sections into which 
the genus has been su os ed, S. Lgnatit ita the only mem- 
ber of the long-tubed sec tion. Ag mi ight be expected, none of 
the Philippine ¢ ten occur outside the group. S. similis shows 
some similarity to S. barbata from New Guinea aad also to 
laurina from Malaya but in the absence of complete material 
the degree of real affinity must remain a matter of conjecture. 
~*® See p. 162 ae Kew Br Bull. 1911, Pp. 300. Since these pages were sent to 
ress, Mr. Merrill, Bota: Bureau of Science, Manilla, has written as 
ollows cams regard to Bichino 08 multiflora, after examining the specimens 
clined to limit the species to Luzon and Mindoro; on account of 
immature fruit. and I believe that additional material will show that this too 
is distinct frown 8. multiflora.. 1 should refer the Clemens specimens from 
Mi a but Micahae Mindanao specimens appear somewhat 
differer T a think that any of the Mindanao specimens should be 
reteryas to 8. maior but ee material will be necessary before the 
