Fig. 13.  Oxalis rosea:  circumnutation of cotyledons, the hypocotyl being secured to a stick; illuminated from above.  Figure here given one-half of original scale.

Another seedling was observed in a similar manner during nearly 24 h., but with the difference that the hypocotyl was left free.  The movement also was less magnified.  Between 8.12 A.M. and 5 P.M. on the 18th, the apex of the cotyledon moved 7 times upwards or downwards (Fig. 14).  The nocturnal sinking movement, which is merely a great increase of one of the diurnal oscillations, commenced about 4 P.M.

Oxalis Valdiviana.—­This species is interesting, as the coty-[page 25] ledons rise perpendicularly upwards at night so as to come into close contact, instead of sinking vertically downwards, as in the case of O. rosea.  A glass filament was fixed to a cotyledon, .17 of an inch in length, and the hypocotyl was left free.  On

Fig. 14.  Oxalis rosea:  conjoint circumnutation of the cotyledons and hypocotyl, traced from 8.12 A.M. on June 18th to 7.30 A.M. 19th.  The apex of the cotyledon stood only 3 3/4 inches from the vertical glass.  Figure here given one-half of original scale.

Fig. 15.  Oxalis Valdiviana:  conjoint circumnutation of a cotyledon and the hypocotyl, traced on vertical glass, during 24 hours.  Figure here given one-half of original scale; seedling illuminated from above.

the first day the seedling was placed too far from the vertical glass; so that the tracing was enormously exaggerated and the movement could not be traced when the cotyledon either rose or sank much; but it was clearly seen that the cotyledons rose thrice and fell twice between 8.15 A.M. and 4.15 P.M.  Early on the following morning (June 19th) the apex of a cotyledon was [page 26] placed only 1 7/8 inch from the vertical glass.  At 6.40 A.M. it stood horizontally; it then fell till 8.35, and then rose.  Altogether in the course of 12 h. it rose thrice and fell thrice, as may be seen in Fig. 15.  The great nocturnal rise of the cotyledons usually commences about 4 or 5 P.M., and on the following morning they are expanded or stand horizontally at about 6.30 A.M.  In the present instance, however, the great nocturnal rise did not commence till 7 P.M.; but this was due to the hypocotyl having from some unknown cause temporarily bent to the left side, as is shown in the tracing.  To ascertain positively that the hypocotyl circumnutated, a mark was placed at 8.15 P.M. behind the two now closed and vertical cotyledons; and the movement of a glass filament fixed upright to the top of the hypocotyl was traced until 10.40 P.M.  During this time it moved from side to side, as well as backwards and forwards, plainly showing circumnutation; but the movement was small in extent.  Therefore Fig. 15 represents fairly well the movements of the cotyledons alone, with the exception of the one great afternoon curvature to the left.