202 TlMEHRI, 
portant physiological questions of wide bearing and 
complex relations that I merely mention in recording 
the information I gathered from collectors and wood- 
cutters during my brief and hasty journey through the 
balata districts, and I must leave them as suggestions for 
investigators more favourably located to attempt to 
settle. It would be interesting to know, too, whether 
there is any difference botanically between what the 
collectors call " man" and " woman" trees — i. e., those 
that yield little or no milk and those that yield much. 
Their own observations are, perhaps, not of much value 
in a question of this kind, but they say there is no differ- 
ence.* The woodcutters I met told me that there is only 
one kind of bullet-tree in the district, but others have 
asserted that there are two, which differ in shape of fruit 
and colour of leaf t Possibly they are forms of the same 
species, like the oval and globose fruited forms of the 
sapodilla, — Achras sapota. 
In the paper by Mr. MELVILLE alluded to in a previous 
page, it is said that the bullet-tree milk is perfectly 
wholesome, and is sometimes used in the "bush" as a 
substitute for cow's milk. Possibly it is occasionally 
* As showing that they are not good observers botanically, I may 
mention that I showed flower and leaf specimens of the bullet-tree to a 
group of woodcutters one day, pressed in my drying papers, not one of 
whom recognised them, or could guess the tree they belonged to. 
f " There are two varieties, differing from each other mainly in the 
shape of the fruit and colour of the leaf. In one, the fruit is oval ; in 
the other it is nearly round, being slightly depressed at the apex, and 
about the size of a large cherry. The leaves are oblong; the upper 
surface is of a brownish tint, and glossy, but in the oval fruited tree the 
shade of brown is much deeper ; the milk also is of a redder hue than 
that procured from the round fruited tree, due to the presence of a 
larger amount of tannin." — D. Melville^ Report to Committee of Corres- 
pondence. 
