78 G. H. KNIBBS. 
general theory of flow in small tubes, which contributed the — 
formule employed in its analysis, leaves little to be desired, that 
of the application of important corrections to the experimental — 
data has been considerably improved, and the whole problem been 
brought more into line with rational mechanics. This, ~ less 
than the fact that former reductions are sometimes not entirely 
free from error of principle, indicates the desirableness of a more 
rigorous analysis of observations by the efflux method, and an 
inquiry as to the most suitable disposition of apparatus when 
accurate results are desired. It will be shown also, that to reach 
the highest possible degree of precision obtainable from expert 
mental data, corrections heretofore ignored must be considered, 
and that the magnitudes of corrections already recognised 
theory, are to some extent uncertain, so that for measurements of 
great precision recourse must be had to methods of observation 
permitting of their elimination. The manner in which this may 
be effected will be illustrated, and a new reduction of the numer- 
ous observations will be undertaken, in which so far as the data 
permit, all the necessary corrections will be made. 
2. History and Literature.—The recognition of the importance 
of a measurement of the viscosity of water seems to have originated 
in the practical requirements of hydraulicians ; the efforts of 
Mariotte,! Guglielmini, Daniel Bernouilli,? Couplet,? D’Alembert,’ 
Bossut,> and Dubuat,® to establish a satisfactory theory of the flow 
of water compelled their attention to the subject. Prony’ expresses 
1 Traité du mouvement des eaux—Paris, 1686, 
* Theoria nova de motu aquarum per canales quécunque fluentes.— 
pes St. Petersbourg, 1726. 
echerches sur le mouvement des eaux.—Hist. Acad. roy. des — 
Poco 1783. 
ssai sur la résistance des fluides, 1752. 
5 Traité élémentaire d’hydrodynamique—Paris, 1775; and Nouvelles 
expériences sur la resistance des fluides, 1777. 
§ Principes @’ hydraulique, verifiés pair un grand nombre d’ expériences+ 
—Paris, 1779. 
7 Recherches physico-mathématiq la théorie d arantes: 
—Paris, 1804. 
