150 
MESSRS. J. N. LANGLEY AND H. M. FLETCHER 
3. The excrertion of the salts injected continued during the whole of the experiment 
—two to three hours. 
4. For a given amount of saliva, the parotid secreted more potassium iodide than 
the sub-maxillary gland. 
5. On injecting potassium ferrocyanide 1 per cent, into the blood, no trace of it 
could at any time be found in the saliva. 
6. On injecting sulphindigotate of soda into the blood, no indigo white could be 
found in the saliva. We tried this because it seemed possible that the known 
absence of sulphindigotate of soda from the salivary secretion, after copious 
injection of it into the blood, might be due to its conversion into indigo-white. 
The Recent Observations of Novi.— After we had completed our experiments 
and nearly finished our account of them, we received a separate cojiy of a paper by 
Novi on the effect of injecting a strong solution of sodium chloride into the blood on 
the percentage of chlorine in saliva. 
Novi’s observations, undertaken at the suggestion of Ludwig, were made on Dogs, 
and in the following manner :—About 80 c.c. of blood were withdrawn from the Dog, 
then saliva collected ; after this a 10 per cent, solution of salt was injected into the 
blood, a fresh sample of saliva collected, and a further portion of blood withdrawn. 
This was repeated one or more times, according to the weight of the animal and the 
freedom of secretion of saliva. The percentage of chlorine in the several samples of 
blood serum and of saliva was then determined. The chief results obtained were : 
1. That an increase of chlorine (as chlorides) in the blood plasma increases the 
percentage of chlorine in the saliva with a given rate of secretion, and may 
increase it with a slower rate of secretion. 
2. That with a rapid rate of secretion an increase of chlorine in the blood plasma 
may increase the percentage of chlorine in the saliva above the maximum that 
can normally be obtained. 
With the former of these results our observations are in the main in aoreement, 
but we find that an increase in the salts in plasma which does not interfere with tlie 
secretory power of the gland has very little effect on the percentage of salts in saliva 
as long as the secretion of water is rapid, and that the increase in the percentage of 
salts which take place with a slower secretion is partly due to the injection of strong 
salt solution leading to a decreased blood ffow through the gland [cf. p. 136, et seq.). 
The latter of these results, our observations leave undecided. For, whilst, on the 
one hand, we never obtained after injection of salts into the blood so high a percentage 
of salts in the saliva as can be obtained normally, on the other, we only once obtained a 
rapid secretion (Experiment 12, IX.) after the injection of salts ; and it is possible that 
in this case the rate of secretion was less than the maximum, and that an insufficient 
amount of salts was injected. 
