AUTHORS' PREFACE xvii 



frequently baffle description and definition. Observation and 

 practice alone can teach any one how to see and recognise them 

 with certainty ; therefore we are fortunate, whenever a variety is 

 distinguished by a constant perceptible feature, to be able to express 

 its distinctness by a single word or a short phrase. Characteristic 

 features of this kind are found in the presence of spines on the 

 leaves of the Prickly Solid Cardoon {Cardon de Tours), in the 

 reversed curve of the pods of the Sabre Pea, in the greenish colour 

 of the flowers of the Dwarf Blue Imperial Pea {Pois Nam Vert 

 Imperial), and similarly in many other cases. 



A part of each description on which we have bestowed mucn 

 attention is that which refers to the seed. In addition to noting 

 the character of its external appearance, we have been careful to 

 state, as precisely as we could, its actual size and relative weight ; 

 and lastly, we mention the length of time during which the 

 germinating power of the seed of each species continues active. 

 It will be easily understood that this could only be expressed in 

 figures representing an average. The duration of the germinating 

 power really depends very much on whether the circumstances 

 under which the seed has been harvested and kept have been more 

 or less favourable. The figures given in this work represent the 

 average taken from an exceedingly great number of trials most 

 carefully carried out. The number of years tabulated is that 

 during which the seeds under trial continued to germinate in a 

 perfectly satisfactory manner. For our present purpose, we have 

 considered seeds deficient in germinating power when they yield 

 only half the percentage of plants which they did in the first year 

 of trial which was made with seeds of the same year's growth. 

 For example, if, in the first year, a certain variety of seeds germi- 

 nated to the extent of 90 per cent., we considered the same seeds 

 to be deficient in germinating power as soon as they began to yield 

 only less than 45 per cent, of plants. Any seeds, of which the 

 germinating power continues active for four or five years on an 

 average, do not entirely lose it after the lapse of ten years or 

 more. It is proper to add that our trials were all made with 

 well-saved seeds. Nothing has a greater tendency to destroy 

 the germinating power of seeds than the influence of dampness 

 and heat. This is what makes carriage through tropical countries 

 so often fatal to their good quality. Up to the present, no 

 better m.ethcd of keeping seeds has been discovered than that 



b 



