708 STARCHES OF IRIDACEiE. 



NOTES ON THE STARCHES OF IRIS. 



Upon the basis of general microscopical characters, these starches appear as two types, one 

 of which is represented by /. florentina, I. pallida speciosa, I. puniila var. cyanea, I. bismarckiana, 

 and I. iherica; and the other by /. xiphium, I. tingilana, I. reticulata, I. histrio, I. alata, and I. 

 caucasica. The first type includes the starches of members of the subgenera Pogoniris, Regelia, 

 and Oncocyclus; and the second, those of the subgenera Xiphion and Juno. Even among the starches 

 of each of these types differences have been recorded in size, regularity or irregularity of outline, 

 fissuration, and lamellation, that are more or less distinctive, as, for instance, as seen in comparing I. 

 florentina, I. pumila var. cyanea, and I. caucasica with the starches of the other irises. There are also 

 in the reactions more or less marked groupings that are in entire accord with the classification of the 

 l)otanist. Thus, in Pogoniris, Regelia, and Oncocyclus the distinctly lower degree of polarization, 

 the less reactivity with safranin, the uniformly higher temperature of gelatinization (an average 

 difference of 6.87) as compared with the members of the other subgenera, is very noticeable. The 

 three starches of the varieties of Xiphium are, on the whole, very close in their reactions; in fact, 

 so close that there can be no mistake about their close botanical relationship. Likewise the starches 

 of the other three members of the subgenus are closely in accord, but bj^ no means so close as in 

 the case of the xiphiums; and their curves differ sufficiently from those of the xiphiums to permit 

 of their differentiation as groups, the differences being noted particularly in the chloral hydrate- 

 iodine, ferric chloride, and Purdy's solution reactions. There is a very close relationship between 

 the members of the subgenus Jiino, and between these and the group consisting of /. tingitana, I. 

 reticulata, and I. histrio, than to any other group. It seems likely that by means of a larger number 

 of reagents and some modifications in the methods that groupings would be brought out that may be 

 strictly in accord with those of Baker; but those of different botanists are not in accord, and it 

 seems that the classification of the entire genus demands considerable revision. 



GENUS MOR/EA. 



Morcea is a genus of bulbous plants closely related to Iris, most of them being popularly Imown 

 as irises. About 60 species are recorded, three-fourths of which are from South Africa, while the others 

 are natives of Tropical Africa, Australia, and Madagascar. They are the southern representative 

 of the true irises of the north. Starch from one species was studied, M. tristis Ker. (Iris tristis). 



STARCH OF MOR.EA TRISTIS. (Plate 67, figs. 399 and 400. Chart 268.) 



Histological Characteristics. In for7n the grains are, as a rule, simple, with occasional com- 

 poimd grains consisting of two components. There are very few clumps. The smaller grains appear 

 to have been in aggregates at one time, as almost all show one or two pressure facets and other dis- 

 tortion from pressure. Some of the larger grains have irregular accretions on their surfaces, so 

 that the form of the original grain may be more or less masked; the outlines tend to be quite irregular. 

 The conspicuous forms are ovoid, either elongated or short and broad. There are also short ellip- 

 soidal, irregularly rounded or quadrilateral, trefoil, rounded triangular, and womb-shaped to pyriform. 

 The small grains comprise round, many sugar-loaf, ovoid, and irregularly polj^gonal. The larger 

 grains do not appear to be flattened. There is a larger proportion of small grains in this starch in 

 comparison with the number found in starches generally. 



The hilum is distinct and comparatively small or of medium size. It is eccentric, so that if the 

 grain is broader than it is long, as many of them are, it lies in the shorter diameter of the grain; 

 it may be double. It is often marked by a small fissure, not deep, and the fissuration may be in the 

 form of a single straight transverse or diagonal line, and usually ragged; sometimes double, triple, 

 or irregularly stellate. 



The lamellce are not distinct. When they can be seen they appear as coarse, continuous rings 

 or segments of rings; often irregular, owing in part to irregularities of the surface, but not always 

 following the outline of the margin. Usually those added last are the most distinct. A broad, 

 non-lamellated space often surrounds the hilum and is bounded by a fairly distinct lamella, within 

 which space the fissures are confined. There are probably 6 to 8 lamellae on a medium-sized grain. 



The grains vary in size from 3 to 30ju; the common size is 20^. The elongated grains average 

 about 20 by 14^, and the rounded quadrilateral about 22 by 24^ in length and breadth. 



Polariscopic Properties. The figiire is eccentric, clear-cut, and distinct in every part, but with 

 many distortions and peculiarities caused by the accretions on the surface of the grain. Portions 



