THE PRODUCTION OF HORTICULTURAL VARIETIES. 



323 



experimental conditions. These would, in the first place, have to exclude 

 the possibility of all free intercrossing, and in the second place, allow the 

 drawing up of accurate pedigrees of the ancestors as well as of the descen- 

 dants of the deviating individual. 



I propose to describe a few cases of such an experimental pro- 

 duction of varieties, taking instances of the sudden appearance of a 

 constant variety, of the working up of an ever-sporting variety, and of the 

 simultaneous production of quite a number of novelties in a single species. 



Linaria vulgaris hemipeloria (fig. 114). — The ordinary toadflax is 

 seen to produce, from time to time, five- spurred flowers. Instead of one 

 spur it has five, instead of four stamens five also, and instead of the two 

 lips of the corolla it has a round, narrow opening surrounded by a wall 

 with five .little lips. Ordinarily, such aberrant flowers are found at the 

 base of the spike, in about one instance among tens of thousands of 

 regular structures. They are easily overlooked, since each flower opens 

 only for a few days and then falls off. 



Linaria vulgaris peloria (fig. 115). — Besides these almost normal 

 cases, from time to time specimens are observed which bear five-spurred 

 flowers only, to the complete exclusion of others. The first instance of 

 this variety noticed was described by Linne from specimens found in 

 1742 near Upsala. This was considered so marvellous a circumstance 

 that he applied to it the term peloria, which is derived from a Greek 

 word meaning " prodigy." The toadflax is a perennial which multiplies 

 itself abundantly by buds produced in great numbers on its roots. In 

 this way all the plants of a small locality may be derived vegetatively 

 from a single specimen. The peloric variety is thus enabled to hold its 

 place during a series of years. After a time, however, it disappears, since 

 it does not produce any seed. In reality, seeds are not wholly absent, 

 but they are so rare that only very few collectors have succeeded in 

 getting any of them. 



There can be no doubt that the peloric variety of Linne had sprung 

 directly from the normal specimens of the ordinary toadflax in its near 

 neighbourhood. In the same way it arose in 1791 near. Gottingen in 

 Germany and about 1825 in the vicinity of Berlin. Of late, it has been 

 observed in 1874 near Zandvoort and in 1896 at Oldenzaal in Holland. 

 Other instances could easily be adduced. They prove that the peloric 

 toadflax is a variety which originates, from time to time and in widely 

 scattered localities, from the common species. 



Starting from these considerations, I have tried to observe its produc- 

 tion directly in order to learn all the details necessary for a thorough 

 understanding of the phenomenon. I cultivated the common species 

 during eight years, giving it all kinds of care, especially excluding the 

 possibility of crossing with allied forms. In the beginning, of course, it 

 did not change, but in the ninth year it produced at once a plant the 

 flowers of which were all peloric, without exception. Since that time my 

 race has reproduced this variety at intervals, partly from the same seed, 

 partly in the succeeding generations. I succeeded in saving some seed 

 from the abnormal specimens, and in the summer of 1898 I had about a 

 hundred flowering plants bearing abundant blossoms, all of which were 

 peloric. 



z 2 



