270 Proceedings of Boy al Society of Edinburgh. [sess. 
water, care being taken to get rid of all air adhering to the weed. 
A well fitting (paraffined) cork was then attached to each of the 
flasks, and through the cork a gas delivery tube passed, which 
dipped into a small mercury pneumatic trough and under an 
inverted test-tube full of mercury. The flasks, with their attach- 
ments, were then left in the laboratory at ordinary (winter) tempera- 
tures. 
After some six weeks the contents of the flask containing the 
ulva and sea water began to evolve gas, and a few days later they 
blackened, while those of the flask containing ulva and tap water 
gave off gas some days later, and no blackening subsequently 
occurred. 
Some of the liquid from the first flask was driven over along 
with the gas, and when the test-tube became full of the latter, the 
liquid escaped on to the surface of the mercury in the pneumatic 
trough. It was found to be strongly acid, and as it evaporated, 
smelt of butyric acid. The gases from this flask were examined 
after an interval of about three months had elapsed since starting 
the experiment, and were found to consist mainly of hydrogen, 
carbonic anhydride, sulphuretted hydrogen, and nitrogen. 
These preliminary experiments gave a distinct clue to the nature 
of the chemical changes which the weed suffers when it rots on 
the foreshore in a moist condition, as well as to the cause of the 
nuisance to which it then gives rise. 
It is clear that an acid is produced in the first stage of the 
fermentation process, while at a later period, and probably by a 
distinct fermentative act, sulphides and sulphuretted hydrogen are 
formed, either by the reduction of the sulphates present in the 
weed itself or in the sea water, or from the albuminoids contained 
in the former, — these sulphides reacting on the iron compounds in 
the tissues of the weed to give ferrous sulphide. The latter 
would no doubt be attacked by the acid, with evolution of sulphur- 
etted hydrogen, and thence the nuisance. As a result of these 
preliminary experiments, we decided to investigate the quantitative 
composition of the gases evolved from the fermenting ulva , and 
also to isolate and identify the butyric acid. 
To obtain the gases, the same arrangement was employed as 
before, only the flasks were placed as soon as charged in an incu- 
