keppel's isle. 65 



instead; an ancliorag*e was marked^ a circumstance 

 which mig'ht have led to serious results^ had w^e run 

 in durino- the nio-ht. 



KeppeFs Isle is from ten to twelve miles in 

 circumference— it is distant from the mahiland 

 six miles. That portion of it seen from our 

 anchorag-e presented rather a pleasant appearance y 

 some fine verdant g-rassy looking* places were^ 

 however^ found on closer inspection to be poor stony 

 or sandy g-round^ thinly covered wdth tufts of coarse 

 g*rass. Behind a long- sandy beach abreast of the 

 ship; an extensive hollow apparently running* back 

 for two or three miles^ flanked by low wooded hills^ 

 was found to be a mangrove sw^amp traversed by 

 several branches of a salt-water creek^ by w^hich the 

 flood-tide g'ains admittance. Here I found numbers 

 of a singailar fish of the genus Chironectes leaping* 

 wdth g'reat activity over the mud among* the arched 

 roots of the mang-roves^ among* which small crabs 

 {Ocypoda and 3Iacrophtkahmis) were making* for their 

 burrow^s in all directions. Fresh water appeared 

 scarce — I came upon one small well^ and beside it 

 a large shell for the purpose of drinking* from. I 

 follow ed the recent tracks of tw o natives^ but they 

 concealed themselves among* the mang-roves^ with 

 their usual caution^ althoug-h armed Avith spears^ 

 as I could see by the marks left during* their hurried 

 flio*ht, and thev knew that I was alone. A small 

 g'roup of women and children Avere afterwards met 

 with by a shooting* party from the ship^ but thej^ 



VOL. I. F 



