270 Mr. Herschel on the 
refinement, may with nearly or quite equal probability of 
success be sought among stars of far inferior magnitudes. 
The proper motions of the stars afford an argument from 
analogy — they bear no relation to their apparent lustre, and 
by far the greatest proper motions known belong to stars 
low in the scale of magnitudes. 
If these remarks have any foundation, it must be obvious 
that we ought not to be deterred from the research of parallax 
by the smallness of the stars composing a double star, or 
their approach to equality. However, we should except here 
stars in, or very near the milky way, below the 7 th or 8 th 
magnitudes, because the probable laminar form of this great 
sidereal stratum affords a presumption almost amounting to 
certainty, that minuteness is here, on the average, an effect of 
distance. But, on the other hand, such large stars as /3 Orionis, 
which are situated in, or near the borders of the milky way, 
with small ones near enough to constitute them double stars, 
have an additional claim to examination, from the additional 
probability thus afforded of being favourably placed for the 
detection of parallax : and moreover, this consideration af- 
fords plausible grounds for a belief, that, in situations remote 
from the milky way, minuteness, on the average, is not the 
effect of distance. 
It is hardly necessary to insist on the great advantages 
presented by the method here proposed, in its complete ex- 
emption from those instrumental errors depending on un- 
steadiness, erronequs graduation, and expansion, and from 
all that uncertainty on the score of refraction, and any doubts 
still remaining as to the magnitudes of the constants of aber- 
ration, nutation, &c., which so much embarrass astronomers. 
