70 Proceedings of the Royal Society of Victoria. 



atmosphere, the results are not very consistent, but it 

 appears to be nearly in an arithmetical progression, Glaisher 

 finding in his 20,000 feet balloon ascent, a decrease of 1° for 

 each 300 feet of ascent. The mean of several observations 

 in Europe and America, is 1° for every 308 feet. 



Humidity is another disturbing element : aqueous vapour 

 has only five-eighths of the density of dry air, and does not 

 permanently retain the gaseous form ; the hygrometer is, 

 therefore, sometimes used in conjunction with the barometer 

 and thermometer in hypsometry, but as the necessary 

 correction is not well established, it is generally neglected. 



The last important source of error is that of gradient. With 

 the air in a state of static equilibrium, two stations at the 

 same height would receive equal pressures ; but as a matter 

 of fact, we find that stations at the same level have at the 

 same moment very different pressures, especially if they are 

 distant from one another. Gradients are the result of such 

 complicated conditions, that there is no way of eliminating 

 their effects. We can only mitigate them by taking 

 extensive series of observations, on the supposition, that 

 the effects will thus balance one another in the mean 

 result. 



Of the many forms of mercurial barometers, I think the 

 Gay Lussac syphon is best suited for mountain work. It is 

 very compact, not liable to derangement, and eliminates the 

 error of capillarity. The zero point is now usually placed in 

 the middle, so that the sum of the upper and lower readings 

 gives the height of the quicksilver. Of aneroids, the 

 Watkin seems likely to become the favourite for professional 

 work, owing to its very open scale, which, in a four inch 

 instrument, has about seven inches of its outer divisions to 

 denote a variation of one inch of mercury. Ordinary 

 travellers, however, will generally avail themselves of the 

 small instruments, of from about one and a half to two 

 inches diameter, which can be easily carried in the waistcoat 

 pocket, where they are generally much less liable to injury 

 than the larger ones carried on a leather strap. It must be 

 mentioned, however, that no system of barometric measuring 

 will give results nearly approaching those obtained by 

 levelling, or even by the measurement of the vertical angles, 

 where the uncertainty of the refraction is a disturbing 

 element. 



From a work issued in 1882 by the Geological Survey 

 Department of the United States, entitled "A New 



