378 PLANTS OF NEW ZEALAND 



Veronica salicifolia (The Willow-leaved Veronica). 



A large shrub ; branches cvhiiclrical. Leaves hnear, 2 in. -6 in. long, shining. 

 Flowers variable, in long racemes, white, mauve, or bluish-purple. Corolla 

 J in. -Jin. diameter. Capsule Jin. long. Both islands, abundant. Fl. 

 Pec. -Jan. JIaori names, Kol-oi-0)ii,iko, Kuroiiiiko. 



Of all the species of the genus, this is probably the most 

 al)iindant. It is almost everywliere common throughout the 

 islands, and along with various hybrids between it and 

 J'rroiiica sjyeciosa and Vcninica iiiacrucarpa, is much cultivated 

 in gardens. There are a nuinl:)er of allied species, which are 

 onlv being gradually separated from it by botanists. The 

 flowers are usually white, and produced in densely packed 

 racemes several inches in lengtli. The leaves are much used 

 as a renredy in cases of diarrh(ca. 



Veponica Travepsii {Travers's Veronica). 



A small shrub. Leaves J in. -1 in. long, jlin.-iin. broad, leathery, flat. 

 Racemes 1 in. -2 in. long. Corolla Jin. diameter. Capsule -Jin. long. South 

 Island: river-beds (Canterbury) ; Southern Alps; Olago. Fl. Jan. -Feb. 



Dr. Cockayne* raised a number of young plants of this 

 species, from seeds gathered from a single parent plant. The 

 seedlings showed marked differences amongst themselves. It 

 is, of course, possible, that this may have been due to 

 hybridization, which, according to Hooker at least, occurs 

 frequently among the veronicas. In order to elinrinate the 

 probability of error from this cause, capsules were collected 

 from three other plants, and their seeds on germination showed 

 the same variability. Xo two seedlings seemed to be alike ; 

 hence Dr. Cockayne came to the conclusion, "that the 

 individual does not nearly [jroduce itself true from seed." 

 Young plants, however, derived from tlie same parent, were 

 more like each other, than they were like the descendants of 

 any other plant. Dr. Cockayne informs us that he is not 

 altogether satisfied with the accuracy of these results, and 

 considers that the experiments should be repeated, with 



•Tiiuis. Vol. XXXI. I). a76. 



