TRANSACTIONS OF SECTION B. 749 
liquid. The microscope, too, shows only a mass of minute spots. If the liquid is 
sufficiently diluted, the nebula does not form till after some minutes. The shape 
of this nebula induced the author to study that part in the fluid, where chemical 
reaction is considerably retarded, which he calls the ‘dead space.’ 
If the solution be poured into a test tube, a space below the surface remains 
clear and transparent. The first thought cannot fail to be, that this phenomenon 
is caused either by sedimentation of the nebula, or by evaporation of the chloro- 
form, or indeed by both. But the shape of the nebula reveals to the observer that 
these cannot be the true causes. If sedimentation were the cause, the nebuia 
would be lowest in the middle ; in evaporation, on the other hand, the same quan- 
tity of vapour would rise from every part of the surface, and thus the shape would 
be different from the one actually seen. For in reality the nebula shows the shape 
of an arch conyex to the surface. 
In order to observe this phenomenon more closely, a glass prism with an acute 
angle was used. Here the appearance was remarkable. Under the surface there 
was a clear space deflected in the direction of the acute angle and corresponding 
to the depth of the meniscus downwards, a space which even inside the angle was 
perfectly clear. 
The ‘dead space’ can be seen in the synthesis of indigo from orthonitrobenz- 
aldehyde, but the reaction is a very rapid one, and for this reason the experiments 
were made with iodic acid and sulphurous acid; 25 grammes iodic acid per litre 
and 0:88 gramme sulphurous acid per litre—the latter being tke solution used by 
Landolt for his time-reactions. A starch-solution demonstrates the presence of free 
iodine. Both in the test tube and in the prism the same phenomena of ‘dead 
space’ can be observed as in the case of the chloral-hydrate reaction. This 
mixture is suited to vessels of the most varied shapes, as shown by experiments of 
a decisive nature. For instance, a tumbler is filled with the reacting mixture, and 
some of the liquid aspirated into a glass tube. The reaction first commences in 
the tumbler, and afterwards in the glass tube, and in its central line only. The 
column of coloured fluid reaches below the meniscus. As in this case the space 
free from reaction cannot be explained either by evaporation or by sedimentation, 
it follows that the glass wall and the surface of the fluid may cause the ‘dead 
space.’ The experiments may be modified by using a tube composed of hollow 
glass bulbs connected by capillary tubes. The reaction is then visible in the 
centre of the bulbs, while the liquid in the connecting capillary tubes remains per- 
fectly clear. The following experiment was made to prove the retardation of 
reaction in capillary tubes. A tumbler and a capillary tube were filled with 
colourless reacting-fluid, and the latter was introduced into the tumbler-fluid. 
After the blue reaction had taken place in the tumbler the capillary tube was 
taken out. Its contents were found to have remained colourless. But when a 
similar capillary tube was filled with the blue fluid, the blue colour could be dis- 
tinctly seen. Tubes which had been blown into bulbs were also filled with the 
‘aad fluid, and it was observed that the reaction began in the centre of the 
ulb. 
An instructive demonstration of the ‘dead space’ and its formation is made by 
fixing a drop of the reacting fluid between the convex faces of two watch-glasses. 
If the reaction be then watched from above, there is nothing extraordinary in the 
fact that the blue colour grows fainter as it approaches the centre, until at last it 
disappears entirely. But if, by means of an apparatus similar to those attached to 
microscope tables, one of the watch-glasses be raised, a sharply defined, colourless, 
and transparent patch is visible in the centre. Moreover, there is a dead space, 
not only in the centre of the drop, but also at its circumference, along the stretched 
surface of the fluid. 
These experiments were variously modified in order to leave no doubt that the 
phenomena were not produced either by evaporation, by sedimentation, or by the 
alkalinity of the glass. 
4 An experiment was then devised in which these three factors were completely 
eliminated by means of the reaction occurring in the reduction of sesquichloride of 
gold with formate of soda. The formate of soda, again, was formed by decom- 
