SUNIER: Marine fish-ponds of Batavia. 267 
definite case found many kepala timah and very many larvae of Myzomyia 
rossii GILES and of a Culex species together in a salt-water ditch with a 
dense floating Enteromorpha vegetation. 
Finally in two cases he was able to observe a number of Haplochilus 
panchax (HAM. BUCH.) in their natural milieu, into which he introduced 
a quantity of mosquito larvae. Only twice did he then see a mosquito larva 
devoured by a kepala timah, whilst generally the fish did not care at all 
about the mosquito larvae, not even when they got right in front of their 
mouths. Nor did the alimentary canal of a few of these kepala timah caught 
among ‘the mosquito larvae put out there contain any of those larvae, but 
only remains of “Daphnia”, “Cyclops” and “Heliozoa”. 
From the preceding SWELLENGREBEL (%) thinks he ought to infer that 
“in salt water not much good is to be expected from the action of fish” 
(scil. Haplochilus panchax (HAM. BUCH.)). 
Now I must begin by pointing out in the first place that the experiments 
and the observations just mentioned do not tally very well together. 
On the strength of experiments in which nothing but mosquito larvae 
were put before recently captured Haplochilus panchax (HAM. BUCH.), 
SWELLENGREBEL (%) concludes that this little fish is “extremely voracious”. 
On the other hand we can only deduce from SWELLENGREBEL’s (%) in- 
terpretation of his observations, that he supposes that (in salt water) the 
kepala timah even when the milieu in which it lives contains a great many 
mosquito larvae, does those larvae little harm, but feeds mainly on other 
small aquatic animals. 
Now it seems to me the height of improbability that a fish in its 
natural haunt should despise another animal for food which is always present 
there in great numbers, but that on the contrary it should devour in cap- 
tivity, great quantities of the same animal immediately after being captured. 
I may also remark the following concerning SWELLENGREBEL’s (°*) 
observations. The simultaneous presence of kepala timah and great numbers 
of mosquito larvae is by no means a conclusive fact, for the possibility 
always remains that but for the presence of the little fish the numbers 
of mosquito larvae in the same breeding place would have been far 
greater still. 
Nor does it seem to me a very telling fact that Haplochilus panchax 
(HAM. BUCH.) was not seen, or rather was only twice seen to devour 
a mosquito larva. To begin with I am in a position to put over against 
this that at my laboratory for more than a year we regularly saw Haplo- 
chilus panchax devouring great quantities of mosquito larvae, in an aqua- 
rium which was in biological equilibrium and in which there were beside 
a tangle of Chaetomorpha, Najas falciculata R. BR. and some Entero- 
morpha, in addition to mosquito larvae and pupae also a great many 
Gammaridea, Copepods, Cladocera, Ostracods, Hydroporines, Sphaerodema’s, 
Nudibranchs etc. etc. In the second place SWELLENGREBEL (%2) does not 
