70 the top sample were fairly good, but the second bottle of the top sample was dirty and offensive, the sediment apparently consisting entirely of fatty matter. I reported that the two samples seemed from the same source by the mineral analysis, but that probably something had got into the one in question at the time of collection, assuming the bottle to have been clean. On inquiry it turned out that to obtain the sample a man had to be sent down the well with a candle, the wallow of which had

more »

... quietly dropped into the well and been carefully bottled ofl' with the sample ! Still another difficulty arises from want of care as regards cleanliness of the vessels in which the samples are sent. I have had some sent in old picklejars, castor-oil bottles, turpentine bottles, physic bottles of all sorts, with corks of a very ancient and fish-like smell. To analyze such samples is simply a waste of time. The best of all examinations is that which is made by the analyst on the spot itself, at the very source of supply. If that be impossible, then the most scrupulous care should be taken about the collecting of the sample and its transmission in clean glass-stoppered bottles, with every detail of information that can possibly be furnished. I should have wished to say a few words about the microscopic examination of water, but my paper has reached the prescribed limits, and I will therefore merely urge the importance of not neglecting this branch of the inquiry. It may be said that it has not as yet done much for us-still it does furnish us with some information ; and if we desire more, it is only by persistent observation and carefully noting what we see and the circumstances under which the objects are seen, that we are likely to extend our knowledge in this direction.