94 
Proceedings of the Royal Society of Edinburgh. [Sess. 
was submitted to twelve hours illumination and twelve hours darkness 
alternately, the numerical data given by Sachs in his Statistical Tables 
( loc . cit.) were used. Those of Polemonium, Calla, and Dahlia were selected ; 
the two former were under observation by Sachs during April, the latter 
in June. The necessary calculations were made on the basis of the natural 
length of the day and night for the particular months, and graphs were 
plotted. The results are given in Plate II., figs. 7, 8, and 9. In spite of 
the short duration of the experiments, which were only carried on for 
about six days, and the consequent paucity of numerical data suitable for 
this purpose, enough is represented to indicate that the phenomenon of 
four-day periodicity is present. In fig. 7, Polemonium, and in fig. 8, Calla, 
there is an obvious four-day increase from 24th-25th to 28th-29th April 
in the former, and from 16th to 19th April in the latter. In fig. 9 the Dahlia 
exhibits part of one of the long eight-day undulations of growth rate 
characteristic of the post-maximal phase of the grand period of an organ 
which has limited growth. 
The four- day rhythm was here represented even when the plants were 
under observation under artificial conditions as to the length of exposure 
to darkness and illumination. 
These results may be regarded as giving some support to the view that 
the long periodicity is to a certain extent autonomic, or based on internal 
changes in the plant, although modifiable by external conditions. Additional 
support is afforded for this view from two series of experiments performed 
on Bean plants at approximately the same stage of growth. In one series 
the whole plant, root as well as shoot, was subjected to normal shoot con- 
ditions, i.e. to the successive alternations of day and night ; in the other 
the whole plant was kept in continuous darkness, that is, stem as well 
as root was subjected to normal root conditions. Results are shown in 
Plate II., figs. 10 and 11, where the four-day waves are equally obvious in 
both cases. 
In Statistical Table, page 98, items 7, 8, and 9 are also of some interest. 
In all three the amount of growth was greater by day, being more than 
double that by night, although in only one, Calla, was this associated with 
a greater mean day rate. In the other two the mean day rates were 
identical and were less than the mean night rates. These results empha- 
sise the necessity of basing conclusions on data derived from experiments 
continued over longer periods of time, in the case of organs of limited 
growth, including the greater part of the grand period, as well as of 
clearly distinguishing growth amount from growth rate by day and night 
respectively. 
