50 



In the Gardens' Bulletin, Vol. T, Xos. 11-12, 1916, pp. 400-405, 

 evidence was produced to show that the Pigeon Orchid flowers 

 about eight days after heavy rain in response to some factor or 

 factors not isolated by experiment. The rainfall was there tabulated 

 for thirty days in advance of thirty flowerings which occurred over 

 the years 1913 to the commencement of 1917. 



In the wet year 1917, the orchid flowered on an unusual num- 

 bei of days, namely. 



January 17th, freely 



„ 18th, Aery sparingly ' 



„ 19th, rather freely 



20th, very sparingly 

 February 19th, sparingly 



„ 25th, very sparingly in the Botanic Gardens, but in great 



abundance in Singapore town 

 March 20th, very abundantly 

 April i;5th, fairly abundantly 



June 2nd, very abundantly 



June 2iitb, one flower only 



July 7tli. rather abundantly 



August 11th, sparingly 



„ 11th, sparingly 



„ 31st, very abundantly, 



September 6th, rather sparingly 



„ 27th, one plant only 



November Sth, very abundantly, 



December 11th, one plant only in the (Jardens, but others in the 

 neighbourhood of the Gardens, 



„ 25th, rather sparingly. 



It is to Ije noted how frequent were these flowerings. Then 

 if tiie reader will turn to the rainfall tables given on pp. 32-33 it 

 can l)e observed tihat the rain which fell on the eighth day before 

 six of the flowerings exceeded an in>cli or on the ninth day before 

 eight of the flowerings, and that in one case only, out of the whole, 

 (lid the rain of the eighth and ninth preceeding days fail to exceed 

 half an inch : that one case heing December 11th, when more than 

 an inch and a lialf fell on the tenth of the preceding da.ys. If we 

 add together tlie raiin in the way in vvhicli it was added on ]). 403 of 

 the first volume we find that on the nineteen days wliich were the 

 tenth before tlie flowering 10.03 inches fell; on the nineteen which 

 were the ninth 23.62 inches; on the nineteen which were the eighth 

 15.17, and on the others in approaching order successively 8.19, 

 6.75, 5.37, 8.97, 7.98, 7.95, and 8.23. Thus the ninth day before is 

 suggested as the critical day rather than the eighth. 



The same general result is obtained if the occasions of abun- 

 dant flowering alone be considered. Per]ia])s higher average tem- 

 ))era.tures lead to tlie eightli day being the critical one: and lower 

 average temperatures to the ninth. 



