144 : The Botanical Gazette. [April, 
119. 1887. P. sf., visit of Plusia—(12) Kerner, Pflanzenleben. 2: 111, 
1891. Protection of pollen.—(13) Peter, Polemoniacee, Engler und 
Prantl, Die nat. Pflanzenfamilien 68: 40-48. 1891. Pollination.—(r4) 
Robertson, Flowers and Insects. Asclepiadaceze—Scrophulariacee, 
Trans. St. Louis Acad. Sci. 5: 578. 1891.—(15) Knuth, Bliitenbiolog- 
ische Herbstbeobachtungen. Bot. Centralblatt 49: 363. 1892. P. acu- 
minata, vis. three butterflies. 
LITHOSPERMUM CANESCENS (Mx.) Lehm.—According to 
Miller (4, 12, 13), L. arvenseis homogamous and regularly self- 
pol-linated, though there is a chance of cross-pollination when 
the flower first opens. According to Kerner (18 Loew 21) 
it is slightly proterogynous, but Miiller says the anthers begin 
to discharge their pollen before the flower opens. JL. purpu- 
reo-coeruleum is slightly proterogynous, with anthers and 
stigma of equal height (17). ZL. arvense has small white 
flowers, rarely with blue (Loew 21) with tubes 4-5™ long. 
Sprengel (1) saw it visited by butterflies, and Miiller (4, 12, 
13) observed as visitors two butterflies, two bees and two 
sytphids. L. purpureo-coeruleum, with red flowers changing 
to blue (17) and tubes 8—9™ long (21), and L. officinale with 
small, dull white flowers, are classed by Loew (14) as bee- 
flowers. In the Berlin Garden the former is visited by Axth- 
ophora pilipes and Osmza aenea, and the latter by Megachile 
willughbtella. 
Bebb (5) discovered that L. longiflorum is only the early 
state of ZL. angustifolium. This and L. canescens are early 
species which are able to attract insects until about the first of 
June (10), when probably on account of being over-shadowed 
by the trees or by the later more luxuriant vegetation, the 
latter goes out of bloom and the former continues to produce 
cleistogamic flowers. Bessey (10) concludes that LZ. angustt- 
folium is not dimorphous, but highly variable, and Halsted 
(16) comes to about the same conclusion. In the case of Z. 
canescens, Smith (9) seems to have regarded the flower as 
dimorphous, but found a rare third form with ‘‘flowers differing 
from the ordinary dimorphous condition.” Bessey (10) Té- 
gards it as a case of well marked dimorphism, though accord- 
ing to Darwin (11) the forms are variable and the case fe- 
quires further investigation. Christy (15) mentions only two 
forms. Halsted (16) calls it decidedly dimorphic, saying he 
has seen no indication of trimorphism. At Madison, Wiscon- 
sin, Trelease (MS. notes) found only two forms and regar 
