ture of garden vegetables must rest upon such a foundation ; but it is 

 still harder to see how we are to avoid such a strait. As a matter of 

 fact, the best horticultural and botanical libraries in this country 

 have been preserving the seed catalogues in recent years, and already 

 we find sundry evidences of their use in some of the most important 

 and dignified of our horticultural publications. Perhaps some of the 

 seedsmen will find it an incentive to the improvement of their publi- 

 cations when they learn that their catalogues are being preserved in 

 the libraries along with Linnaeus' Species Plantarum and Louden' s 

 Encyclopedia of Agriculture. 



There has been much discussion of late as to the value of variety 

 testing in the experiment stations, and the tendency has been to dis- 

 parage the worth of comparative tests of long lists of garden varieties. 

 Without breaking into that debate, it may be pertinent to suggest 

 that, if such tests could result in accurate descriptions of varieties, 

 with careful and exhaustive study of the proprieties of nomenclature 

 in each case, a real and permanent value would be added to them. 

 Some work of this sort has been done already enough to point the 

 way. Hardly more than that has been accomplished in the nomen- 

 clature of fruits. But these beginnings are very interesting, and, 

 considered as suggestions for future work, they are invaluable. I 

 will call attention to some of them in the next chapter. 



