J 
| figs. 1 and 2). 
INSECT POWDER 25 
DALMATIAN FLOWERS 
1. Numerous 3-pored, spiny pollen grains (PI. III, figs. 3 and 4). 
2. Marginal papiliz and papille in surface view presenting a 
puckered” or 3-cornered appearance (PI. I, figs. 1 and 2). 
3. Sinuous, striated cells of epidermis of ligulate florets (Pl. I, 
14 
to} 
4. Toothed fragments of the disk florets (Pl. III, fig. 2). 
5. Shredded fragments of involucral scales, strongly lignifed (PI. I, 
pam. 5) 
6. Attenuated and twisted horizontal cells of T-shaped hairs from 
_ the bracts (PI. II, figs. 3 and 4). 
7. Occasional strands of stem tissue, much larger than other frag- 
_ments, and usually possessing roughened or fibrous ends (PI. I, fig. 6). 
8. Oil glands from the corolla and fruit. These are very seldom 
detected in the powder. 
9. Somewhat rectangular patches of sclerenchyma tissue from the 
fruit, containing numerous diamond-shaped crystals exhibiting a va- 
riety of colors under polarized light. Numerous isolated stone cells 
are also often found in powder ground from mature flower heads 
(Pi, fies..3 and.4;° Pl. 111, fig. 6). 
PERSIAN FLOWERS 
The diagnostic characters of the Caucasian or Persian flowers 
(Chrysanthemum roseum) in the powdered form are rather similar to 
those of the Dalmatian flowers. As already stated, the papille of 
| the ray florets differ somewhat, and the achene tissues do not contain 
| the crystals characteristic of the Dalmatian flower fruit. The reddish- 
| brown scarious margins of the bracts are often more striking in the 
| Persian powder than in the Dalmatian. Collin (47), who also studied 
the Persian flowers, has summarized the principal diagnostic char- 
| acters which distinguish them from the Dalmatian flowers. 
SUMMARY 
Of course, the relative abundance of many of the tissues mentioned 
depends upon whether the powder is ground from open or closed 
| flowers. Pollen is more abundant in closed-flower powder and 
_sclerenchyma, tissues in open-flower powder. The greater part of 
the pollen of closed or immature flower heads still remains in the 
| closed “buds” or heads, the pollen not having as yet been scattered 
| by the wind. On the other hand, the mature flower heads are prac- 
| tically devoid of any great quantity of pollen, but contain the mature 
| achenes, or fruits, still attached to the receptacle, or, very often, 
| fallen out, depending upon the ripeness of the heads. Consequently 
| appreciably little of the achene or fruit tissues is found in closed- 
flower powder, whereas powder ground from open flowers is rich in 
the lignified tissues of the fruit and contains no pollen grains. 
Stem tissues occur in all powders to some extent, although they 
| should not be present in excessive quantity. Careful study of the 
| stem and flower head tissues shows that there is not the shghtest 
| difficulty in distinguishing between them. 
The following references in the bibliography (p. 78) deal with this 
| subject: 23, 24, 29, 47, 69, 106, 115, 116, 117, 118, 119, 187, 148, 150, 
| 152, 156, 162, 163, 169, 174, 186, 199, 202, 246, 249, 257, 258, 261, 
| 278, 279, 281, 282, 285, 289. 
