156 Some Factors in Plant Competition 



surprising results. According to the usual conception of things dense 

 overcrowding produces thin, unhealthy plants, badly drawn up, with few 

 and small leaves often of a bad colour, whereas plants with plenty of 

 space become stout and stocky, with a large area of healthy leaves. 

 In the present instance overcrowding was apparently favourable to 

 growth for some long period, as the plants became so much larger, with 

 very broad leaves of an intensely dark colour, while the corresponding 

 spaced plants were considerably smaller, with narrow pale leaves, which 

 tended to die off at a rather early stage. The reason for this reversal of 

 the usual order of things is not obvious, but certain suggestions may 

 ultimately throw some light upon it. The experiments were carried on 

 in a roof greenhouse which receives plenty of sun and tends to become 

 very hot at times. The range of temperature during the course of the 

 experiment is shown in Table VII. 



Strong sunlight tends to check growth to some extent, and as the 

 spaced plants were each so far separated from their neighbours as to 

 permit of no shading, the insolation had full play and possibly kept 

 back leaf development. The strong light may also have had some effect 

 upon the chlorophyll, preventing it from attaining a very deep colour. 

 The crowded plants cast considerable shade on one another thereby 

 reducing the insolation, and this may have enabled the plants to form 

 longer and broader leaves as growth was less checked by the light, 

 while the chlorophyll developed a very full deep colour. The amount 

 of shading, however, did not seem to be sufficient to account for the 

 marked difference in external appearance, and other factors were sought 

 for. The wet and dry bulb thermometers which were kept in the house 

 and among the crowded leaves showed that except on rare occasions 

 the humidity of the air among the crowded plants was greater than that 

 in the free air of the house. The range of variation was very considerable, 

 reaching to as much as 50 per cent, saturation, as is shown in Table Vlll. 



The character of the crowded leaves suggests a correlation with this 

 difference of humidity. The succulence and slight fleshiness of these 

 leaves recalls those of plants growing in very moist places, in marshes or 



