OBITUARY NOTICES. 
451 
period may be due to an orbital motion by reason of which 
the component velocity, if directed toward the earth, has a 
steady acceleration or, if directed from the earth, has a 
steady retardation, and that we may expect the observed 
shortening to eventually cease and be followed by a length¬ 
ening of period. This hypothesis has been lately and in¬ 
dependently renewed by Chandler, and has been verified 
by spectroscopic observations of this star. 
(6.) In 1856 Ferrel communicated a memoir to Gould’s 
Astronomical Journal on the problem of the “ Tides with 
regard to oscillations of the second kind,” a problem to 
which he returned in his “Tidal Researces of 1874,” and 
again in a paper in the Philosophical Magazine of 1876, 
and finally again in Gould’s Astronomical Journal for July, 
1889, and December, 1890. In these various articles Ferrel 
maintains that an error of principle pervades the deter¬ 
mination of a certain constant by Laplace, and the attain¬ 
ment of certain results as first promulgated by Laplace and 
accepted by Airy, Bowditch, Thomson, Darwin, and other 
mathematicians. He corrects both Laplace and Airy, and 
contends in effect that in an ocean of uniform depth around 
the earth and composed of fluid having no viscosity the 
semi-diurnal tidal wave ma}^, for one critical depth only, 
have any amount depending upon certain initial conditions, 
but that for all other depths and for the case of nature, 
where friction exists, the initial tidal motions are gradually 
annulled, so that the tides observed by us are essentially 
those forced tides that are directly maintained by the sun 
and moon and the free tides depending on these latter, but 
not those that depend on initial conditions. 
(7.) 1856.—In a so-called exposition of the phenomena of 
the gyroscope Ferrel works out the now well-known rela¬ 
tions between the composition of rotations around different 
axes. Of course many others were at work upon that prob¬ 
lem at that time, stimulated by the experiments of Foucault, 
but Ferrel’s solution was entirely independent of European 
investigators, and was, I think, the first that was published 
