THE INFLUENCE OF SALT AND OTHER SOLUTIONS ON THE FROG. 177 
It may be mentioned here that, in view of the possible objection that 
this effect might be due to bacterial toxines, this experiment was repeated 
under special conditions. The embryos were kept in a constant stream of 
sugar solution, which flowed through the apparatus at the rate of two 
litres per diem. 
D. The fourth class comprises urea and sodium sulphate ; in these 
development is nearly (urea) or quite (sodium sulphate) normal in form 
and rate. 
Little remains io be added to the account given last year, except that 
the microscopical investigation has fully borne out the statement made 
of the external characters of these embryos, 
To sum up, these isotoni¢ solutions may be grouped according to their 
effects as follows :— 
1. Solutions which kill the egg in an early stage (segmentation or 
gastrulation). 
2. Solutions which kill the embryo at a rather later stage, when the 
medullary plate is being formed, without permitting any very great degree 
of differentiation. 
3. Solutions which, though they allow differentiation to proceed for 
some way, do distort development :— 
a. The embryo remains spherical. 
B. The embryo elongates. 
Differentiation may go as far in a as in /3, 
4, Dextrose must be placed in a class by itself ; it seems to affect the 
rate only, the form of development hardly at all. 
5, Solutions in which development is nearly or quite normal. 
It is clear that it would be extremely difficult to assign the whole of 
the effect produced in each case to the increased osmotic pressure of the 
medium, and to that alone ; to do this it would be necessary to assume 
that the eflect was inversely proportional to the permeability of the 
embryo to the substance employed. There is, of course, no direct evidence 
for this whatever. 
Animal tissues are, indeed, supposed to be more or less impermeable to 
mmagnesiuni salts, which produce a less, and permeable to sodium chloride, 
which produces a greater, effect. 
The toxicity of the reagents must far more probably be set down to 
some other physical cr chemical property they possess, though what this 
is it is impossible to say exactly at present. This is clearly shown by an 
experiment which has been made during the present year. 
EE: 
Ammonium bromide is one of the most poisonous of the substances 
‘tried ; in a solution whose concentration (1-04 per cent.) is isotonic with 
that of a 0:625 per cent. sodium chloride solution the egg dies during 
segmentation. 
Weaker solutions have now been tried, and this is the result :— 
0-26 per cent. : a segmentation cavity is formed ; then a dorsal lip of 
the blastopore and a very short archenteron ; death follows. 
0°13 per cent. : the result is the same. 
0-065 per cent. : the result is the same. 
1905, N 
