200 RELATION OP SITUATION TO COLD. 



heat or to drain, and they will, therefore, he the most unfavour- 

 ahle to the operations of the gardener ; extremely light sandy 

 soils, on the other hand, part with their moisture so rapidly, 

 and absorb so much heat, that they are equally unfavourable. 

 It is the light loamy soils, which are intermediate between the 

 two extremes, that, as is weU known, form the best soil for a 

 garden. Situation is, however, of more consequence than soil, 

 for the latter may be changed or improved, but a bad (that is, 

 cold) situation is incurable. Cold air is heavier than warm air, 

 and, consequently, the stratum of the atmosphere next the soil 

 wiU. be in general colder than that above it. "When, therefore, 

 a garden is placed upon the level ground of the bottom of a 

 vaUey, whatever cold air is formed upon its surface remains 

 there, and surrounds the herbage : and moreover, the cold air 

 that is formed upon the sides of low hills rolls down into the 

 vaUey as quickly as it is formed. Hence the fact which to 

 many seems surprising, that what are called sheltered places 

 are in spring and autumn the coldest. We aU know that the 

 Dahlias, Potatoes, and Kidney-beans of the sheltered gardens 

 in the valley of the Thames, are killed in the autumn by frosts 

 whose effects are unfelt on the low hiUs of Surrey and 

 Middlesex. Daniell says he has seen a difference of 30°, on 

 the same night, between two thermometers, placed the one in a 

 valley, and the other on a gentle emiuence, in favour of the 

 latter. Hence, he justly observes, the advantages of placing a 

 garden upon a gentle slope must be apparent; "a running 

 stream at its foot would secure the further benefit of a contigu-. 

 ous surface not liable to refrigeration, and would prevent any 

 injurious stagnation of the air." 



One of our German translators has expressed his opinion that no such 

 difference as 30° can have been observed, and alters the statement to 3° 

 Reaumur! But if he had consulted Darnell's Meteorological Essays, Ed. 

 2. p. 525, he would have found that the quotation is exact. No doubt it 

 was an extreme case ; but Mr. Thompson remarked last April that at 

 the time when fruit-buds near the ground had been universally kUled by 

 , 14° of frost, they were safe on trees twenty-five to thirty feet above the 

 level, and he beUeves there may have been a difference of 10° — 12° in 

 favour of even that slight height. 

 As a good example of the practical mode of dealing with low 



