THE IMPKOVEMliNT OF CARNATIONS. 155 



little work in melons, and in the melon work my experience was about the same as that 

 of Professor Bailey — but I found in working carnations that where you inbreed, say you 

 cross crimsons, take Maceo and fertilize Maceo, and you follow that up to the extent of 

 making the foliage almost crimson, as soon as you got to that point there wasn't any- 

 thing left in the constitution of the plant; it was run out, in other words. I didn't get 

 any seed to speak of from it, and what seedlings I did get were runts. Now, I exercise a 

 certain amount of selection in my seedlings. That is, after they show the fourth or fifth 

 leaf, if the habit of the little seedling doesn't suit me I pull it up and throw it away. 

 Seedlings of a certain habit are of no value; and after three or four years of work I found 

 that out, so we don't waste any more work on them. But in carnations I found it nec- 

 ecessary, where working within very close lines, to introduce influences from the 

 outside; that is I take, say, a single crimson with a strong habit, and I will breed into 

 that single flower in order to get strong habit, and then go on with that; and I do the 

 same with whites and scarlets, and all the way through. 



H. H. GrofT: Allow me to report an experience with a double or semi-double type 

 of gladiolus, the flowers of which when separated and placed in a shallow vessel of water 

 resemble the Xyniphaea. Usually that blood in many of my crosses has resulted in the 

 preparation of seeds producing twins, and in one case, in order to prove the question of 

 these being identical, I kept the first ones separate and found that they produced identical 

 flowers. Since then, my last season results show tne largest number of twins from single 

 seeds that I have ever produced within my experience. 



C. \V. Ward: I have heard that some hybridizers advocate keeping no records. I 

 have religiously kept a record of every cross I made since 1894. Now I find those records 

 valuable in this way: I don't study them very much when I make my crosses, but after 

 I have accomplished a resu - I can go back and find out how I did it, and there is the 

 most value. Now I can trace every single crimson flower there back to Maceo or Gomez; 

 and when it comes to the whites or any particular stripe, in tracing the pedigree you wiT, 

 find that from the pedigree in both parents the color that you have had exists some- 

 where back there. Now, when you find that you make a cross, for instance, between two 

 yellows and you have got ten scarlets and three yellows, and if you go back you will 

 find that the scarlet has been the predominating color in the pedigree on both sides. So 

 that in just studying my pedigrees I have come to the conclusion that the closer you 

 breed to a color the surer you are of getting that color; and that also holds good for 

 fragrance. 



H. H. Grofi': I also kept notes until after the production of over 5,000 desirable 

 types, and it became rather burdensome. 



C. W. Ward: I also produced a large numlier of desirable forms, but then I picked 

 the best of the desirable forms. 



W. Bateson: I would like to know whether in breeding from the light sections you 

 ever produced dark seed. Breeding from the dark sections you get to the light, but 

 breeding from light sections, question, do the darks ever come? 



C. W. \V'ard: In pinks they uo; we get dark pinks. But I never got any crimsons 

 e.xcept in dark pinks where the crimson existed in the pedigree. I presume that this Day- 

 break pink that I get in the crimson comes in from Mr. Dorner's work; I can't trace it in 

 mine. 



W. J. Spillman: Careful work of this kind is of immense value to the scientist, and 

 I can't help saying either that when these practical men, who occasionally rather speak 

 slightingly of theoretical men, learn the immense value in dollars and cents of knowing 

 the why and the wherefore, and how to do it again, it will be worth a great deal to them. 

 In other words, if Mendel's law is true, it is worth millions of dollars to the breeders of 

 plants in this country. If it is not true, it is vastly important that we should know it soon. 



