18 XOTES ON THE VKGETATIOX OF NEV,' GUINEA. 



slender stems 1 ^ft. to 2ft. high. Scape shorter than the stem, 

 bearing an elongated head 2 or 3 inches long, the broad imbri- 

 cate bracts with a pink or blood- red margin at the upper end 

 and a subulate point in the centre. Ehizomes very pungent 

 and not very fibrous. Stems of sugar-cane wei'e generally 

 brought out in the canoes, but this was the only place where 

 ginger was offered for barter. No canoes are ever without a 

 supply of cocoanuts, Cocos nucifera, Linn. These and a dirty 

 kind of sago and fish form the principal food of the natives. 

 There are a number of varieties of the cocoanut, some differing 

 in shape of fruits, others in the growth of the tree. This seems 

 an apt opportunity for making a few remarks on cocoanut 

 planting. I know that it is a pretty general belief that these 

 trees will thrive on coast sands without fresh water — or, as it 

 might be, brackish water — within reach of their roots. This, in 

 my opinion, is decidedly an error, as very few plants will thrive 

 tinder such conditions, and certainly not the cocoanut palm. 

 When to the observer's eye such may appear to be the case, it 

 will usually, if indeed not always, be found that there is an 

 undercurrent, or a strata of the sandy soil saturated with fresh 

 or brackish water, or that there is a frecpient rainfall in the 

 locality. With regard to planting the nuts, also, great care 

 should be taken never to place them deep in the soil. I would 

 recommend placing the nuts lengthwise, sunk in the soil for 

 about half their transverse diameter, and to prevent them moving 

 three 'or more stakes might be driven around each nut. Of 

 course if the nuts have sprouted n nd formed roots before planting 

 these latter should be spread out horizontally and covered with 

 the sand or soil. Near one settlement we were taken to see a 

 plantation of these trees which were not thriving satisfactorily. 

 The cause was plainly due to bad planting, shallowness of sand 

 over the coral, and therefore a scanty supply of fresh or brackish 

 water. The sand had been removed down to the coral bed, say 

 about Hft. to 2ft., and the nuts placed upon the coral and 

 covered with sand. The whole of the locality was low, with no 

 sandy ridges where the rainfall could be stored up to supply the 

 low-lying parts in time of need. From this plantation we were 

 taken a short distance to see a clump of these trees near a hut, 



