226 



THE AGRICULTURAL NEWS. 



July 15, 1916. 



As most people know, the germination capacity of 

 seeds can easily be tested by counting out a certain 

 number and keeping them on damp blotting paper, care 

 being taken to allow the presence of fresh air. There 

 are one or two aspects of this apparently simple test 

 that require careful consideration, however. It is 

 necessary first of all to distinguish between germination 

 capacity and rate of germination. One may have a 

 hundred seeds all of which will have germinated 

 uniformly in three days, while the germination of 

 ■another hundred of the same kind may be spread over 

 a week. The first sample will obviously give a more 

 uniform and an earlier crop of seedlings than the second. 

 Yet the ultimate germination capacity of each is the 

 same. 



Some onion seed which had been kept in cold 

 -storage for a year was recently tested by this Depart- 

 ment with interesting results. The method adopted 

 was not the orthodox one of taking one lot of 

 a hundred; three lots of a hundred each were counted, 

 -and these divided into groups of ten. 



The following were the results:- 

 In first groups In second groups 

 often, often, 



2 germinated. 2 germinated. 



•3 „ 3 „ 



3 

 3 

 4 

 5 

 5 



6 

 (5 

 8 



In third groups 

 of ten, 

 3 germinated. 



3 „ 

 3 „ 



1 „ 



3 „ 



2 „ 



4 „ 



9 



5 „ 



48 per cent. 42 per cent. 28 per cent. 



It will be seen that the germination was very 

 uneven, and that the percentage germination figure is 

 absolutely unreliable. That this is so can be seen by 

 examining the figures in the first column, where there is 

 an extreme range of from 2 to 8. As a matter of fact 

 the second two experiments were merely conducted as 

 confirmatory tests of the inference, that the 48 per 

 cent, is undependable. The point to be emphasized is, 

 that taking one group of a humlreil out of a considerable 

 quantity of seed is not sufficient to give reli.able results. 

 It is the same principle that holds good in the 

 case of manurial and other plot experiments: 

 duplication is necessary as a means of exposing 

 the degree of variation. It must be pointed out, 

 however, that the percentage figure for germination 

 becomes more and more dependable the nearer it 

 approaches 100, and conveisely, the nearer it approaches 



0. So that a germination percentage of, say, 9.5 may be 

 accepted as being practically reliable. 



Good onion seed has a germination of between 

 90 and 99 per cent. Cotton has a germination of 

 80 to 90 per. cent. But where so-called 'bad germina- 

 tion' occurs, it is by no means always the fault of the 

 seed. 



In the West Indies, small seed like that of 

 most vegetables is frequently carried away by ants. 

 This was probably the cause of the failure referred to 

 by our correspondent. In gardening work the visits of 

 ants may be frustrated by placing the seed boxes on a 

 bench the legs of which rest in small receptacles 

 containing water. In the garden itself the most satis- 

 factory method is to locate the nests, and kill the 

 colony by application of carbon bisulphide. Birds may 

 ■ also be responsible for the loss of planted seed. This 

 menace can be circumvented by employing wire netting. 



Another factor which may interfere with ger- 

 mination IS a badly regulated water-supply. Too 

 much water is as bad as too little. Though not 

 strictly connected with the actiuxl process of ger- 

 mination, it may be noted that 'damping off" of 

 seedlings is closely related to exce.ssive moisture condi- 

 tions and overcrowding. Lastly, seeds should not be 

 planted too deeply, and should be sown only in well 

 tilled soil in fine condition. This latter remark 

 applies to the field cultivation of cotton. Cases are 

 known where coarse pen manure applied to the land 

 too soon before planting has led to bad germination 

 resulting in an uneven crop of cotton, and conse- 

 quently an unsatisfactory one. 



SUGAR INDUSTRY. 



SUGAR-CANE EXPERIMENTS IN BRITISH 

 GUIANA. 



A compressed account of the re.sult.-^ of important experi- 

 ments conducted in connexion with .sugar-cane in British 

 Guiana by the Department of Science and Agriculture during 

 191-1-15 appears in a recently issued Department Report. 



In this it is stated that work in connexion with the 

 raising of seedlings of sugar-cane was continued. In all 

 9,730 seedlings were raised, and of these 3,510 have 

 been retained at the Botanic Gardens and 1,517 distributed 

 to sugar estates in the Colony. Of the seedlings retained at 

 the Botanic Gardens, 1,486 were po.ssible crosses, while 31-5 

 were s-lfs. A list is given of the more important varieties 

 crossed. 



The general crop returns in the experiments were not 

 very satisfactory, due in part to the defective rainfall, but 

 more largely to the long period the various tieUls have been 

 under sugar-cane cultivation without rest. Nevertheless the 

 accumulated result.s obtained are significant and of great 

 interest. 



