1887.] of the Horopter. 65 
therefore 
@ sin y _ ycosy Zz (5) 
2 2 ar 2 2 = 5 - p @8eeoes 9) 
a, +a, —2a,a,cosd a,—a, 2a,a,singd 
These values of x, y, z are proportional to the direction cosines 
of the tangent to the horopter curve. 
They shew that when a, exceeds a,, and ¢ is measured round 
in such direction that it is less than two right angles, the horopter 
lies in the quadrant of the field of vision which ranges from a, in 
the positive direction, and that it slopes away from the eyes in that 
quadrant. 
Further approximation in equations (1), (2), (3) would lead 
easily to the determination of its curvature, but the results are 
too complicated to be of much use. 
9. When the point of vision is in the medial plane, a,=a,, 
and the direction of the horopter is in the medial plane, inclined 
to the plane of vision at an angle whose tangent is 
sin ry cot 5 
and it slopes away from the eyes in the upward direction. 
When ¢ = 0, the direction is in the plane of vision, and makes 
an angle with the axis of 2 whose tangent is 
a, — a, 
These are in agreement with known results: in the first case 
the arc in question is part of a line, in the second it is part of 
Miiller’s circle. 
(3) On the finer structure of the walls of the endosperm cells 
of Tamus communis. By WALTER GARDINER, M.A. 
It would appear from the author’s more recent researches that 
the perforation of the walls of the endosperm cells in the plant 
referred to, is established after the formation of the wall, and in 
a similar manner to that which occurs in sieve-tubes during the 
formation of the sieve-plate. The author further hopes to shew 
that this is a special instance of a phenomenon, 
