39 ^ Mr. de Luc on 
only 3 grams was not fufficient to keep their index fteady, either 
in water or any where; and this he obferved himfelf. 
74. Notwithftanding that natural explanation of the unftea- 
dinefs of M. db Saussure’s firft hygrometer when in water , I 
thought it neceffary to try, in the fame circumftance, fome 
well-conftru&ed hair hygrometer ; therefore I made two, fimi- 
lar to the laft of M. de Saussure’s in every refpedl, except 
that of having their index at the top; and for the connection 
of the hair with the axis , I ufed a kind of pincers , not above 
half a grain in weight, with a hair- like bit of hemp, which I 
know does not alter in any fenfible manner the march of the 
inftrument. Now, thefe hair hygrometers, with their weight 
of only 3 grains , being put into water , follow in it their own 
laws, arrive and remain fixed at their once fettled point, as 
well as any of my other hygrometers. This M. de Saus- 
sure, would have alfo found, if in the improvement of his 
inftrument the axis had remained at the top. 
75. Before I explain the influence which that accidental cir. 
curnftance of the place of his index has had on his opinions with 
refpedt to extreme moijlure , I muft mention another of the 
fame kind which has contributed to the fame effe£h It is evi- 
dent, both from theory and from M. de Saussure’s own expe- 
riments related above (§ 66.), that a fufficient quantity of wa- 
ter in any part of a clofe veffel is the only requifite for pro- 
ducing in it the maximum of evaporation ; but, with a view of 
accelerating that effedt for the common purpofe of fixing the 
point of extreme . moijlure on his hygrometer , he prefcribes 
wetting the infide of the veffel, befides inverting it over 
water . He did not (and indeed he could not) forefee the 
confequence of that alteration in his firft procefs ; but in 
fadl it was fuch as to prevent him from difcovering, even with 
time. 
