A Route for the Transfer of Materials Through the Pollen Grain Wall

In the course of a study of pollen wall development in the Gramineae two interesting relationships were noted: the locule wall and tapetal cells were bordered by spheroids which were similar chemically (resistant to acetolysis) and morphologically to the pollen exine. The pollen exine as well as the wall of the spheroids was perforated by submicroscopic channels. In the mature anther the spheroids averaged 0.45 # in diameter without spines and had a central cavity measuring 0.2 # (Fig. 4). Their surface was covered uniformly with small spines of the same size and shape as the spines on the grass pollen ektexine. The spheroids were located between the nutritive surface of the locule and the pollen grains (Fig. 2). The only exception to this positioning is that they were sometimes seen between adjacent tapetal cells. Where tapetal cells extended into the locule as in Fig. 2, these bodies followed the tapetal margin into the locule. During the early microspore stages when the exine was thin, the spheroids were small, they had only a thin outer spinose ring of a density like the microspore exine, and the central area had an intermediate electron density. The pollen wall developed rapidly prior to microspore mitosis, and the size of the spheroids and thickness of their exine-like wall also increased until by microspore mitosis their dimensions were maximal. At this stage the central cavity had a low density similar to the background methacrylate (Fig. 4). When microspore development was most intensive (intermediate between telophase II and microspore mitosis), cytoplasmic strands could be followed from the tapetum directly to the exine or from the tapetum to the spheroids and from there to the exine. After the intine matrix was formed (prior to microspore mitosis) giving a transparent and contrast-abetting region inside the exine, cytoplasmic strands could be traced from the pollen protoplast through the intine (Fig. 4) and lacunae of the ektexine. In 1952 Dahl (1) reported strands of dense material in the intine of Tradescantia paludosa. Thus the spheroids can be considered as part of the connections between the male gametophyte and its supporting tissue. Their proximity is not, however, necessary for exine building since sphe-roids were not found between adjacent microspore walls (Fig. 2) nor on the microspore surface toward the locule interior, though exine development here equalled that in direct contact with the tapetum and the zone of the spheroids. The number …