August 3, 1918 



HORTICULTURE 



101 



Sports 



Iris in a garden which has been prepared as directed 

 in this letter is sure to sport, and that is one reason why 

 the land was made so rich. The present arrangement 

 of the parts of an iris flower of the type with erect seg- 

 ments to the perianth is so perfectly beautiful, it can- 

 not be improved. If therefore the sport results 

 in any change in the position or number of 

 the parts of the perianth, using this term in 

 its strict sense, it should be discarded. Some 

 kinds of iris have a considerable tendency to thus sport, 

 and where the ovary shows signs of having more than 

 three divisions, and there are more than three 

 stigmas it is possible to get a new type with 

 more segments to the perianth, and at length to get this 

 type to breed true. In my garden this has been true 

 more frequently with the form named Kharput than 

 with any other, I imagine, therefore that this variety 

 is in what de Vries calls the mutating stage, or is 

 brought to this condition by high cultivation; a muta- 

 tion plant is one that gets ehil(]ren that differ in more 

 than one feature from the type and breeds true. 



The Third Step 



consists of writing on a painted wooden label, the date, 

 the names of the parents, or a sign to indicate this, and 

 the number the flower is to have in the record book. 

 Then attach the label to that particular flower, and 

 make the record. Before leaving the flower, strip back 

 the bud sheatli — B S — Figure 5 — to prevent the water 

 it will hold after a rain from injuring the developing 

 seed pod. Then pass on, leaving the fourth operation 

 to the plant. 



The Fourth Operation 

 This you delegate to the flower. Soon after a vigor- 

 ous pollen grain touches the sensitive surface of a 

 stigma it begins to develop a tube which, lengthening, 

 grows down through the tissue of the stigma until it 

 reaches an ovule — shown in Figure 3 — as one of the 

 Avhite dots in the ovary 0. The end of the tube then 

 opens and out come two bodies, the sperms, one of which 

 xmites with one part of the ovule and the other with 

 another. This is the double fertilization of which you 

 can find an account in the recent books on botany. 

 When complete a new individual has arisen. In the 

 case of a cross between plants having differences, the 

 plant when developed will have some of the characteris- 

 tics of each parent and the abihty to transmit these, 

 tbough not always in the same combinations. Generally, 

 with the common garden iris tlie results will be differ- 

 ent from what you expected. Years ago I bought 

 Pallida speciosa. From its name you would have 

 thought it was surely the type of a true botanical spe- 

 cies. Would you not have supposed Pallida Dalmatica 

 was a true species from Dalmatia, and PaUida Austra- 

 lis a true species from Austraha? In my garden they 

 were not any of them true botanical species, as they did 

 not breed true when protected from Vicinists, that is 

 other varieties of iris. Beside Pallida, iris comes to us 

 under such apparently species names as Junonia (an- 

 other name at one time for Pallida), Sambucina, Jac- 



quiniana, Trojana, Cypriana, causing the beginner to 

 suppose that if he raises seed from these he will get 

 plants like the parent ; but if his results are like mine 

 he will get a series of severe jolts. They will show him 

 what a long and at times weary journey must be taken 

 to learn iris genealogv'. Yet this must be pursued to 

 the end before crossing can be on a scientific basis. 

 Science is only a technical name for the collected re- 

 sults of accurate observations, and we gardeners become 

 scientific men as soon as we observe accurately 



A Typical French Iris 



In Figure 1 is shown a Lily of France of the form 

 we should strive to reach and perpetuate. It belongs 

 to a group to which I have given the name Athena. 

 The flower resembles that of Pallida Dalmatica, but 

 the last trace of purple that injured the drooping seg- 

 ments of this flower has been nred out leaving the whole 

 flower a clear uniform pale lavender blue, the Bleu 

 Lavande, Xo. 204 of the Repertoire de Couleurs of the 

 Societe des Francaise des Chrysanthemistes ; though 

 the shade is lighter than No. one. The flowers are 

 nearer together than in Dalmatica, making the stalk 

 more impressive. Like most of the other species de- 

 scribed in papers classifying iris, this group does not 

 form a true botanical species. The name was chosen to 

 represent a new horticultural group, and used with this 

 understanding is convenient. 



