Ultrastructural interpretation of the Late Cretaceous megaspore Glomerisporites pupus and its associated microspores

Abstract

The ultrastructure of the Late Cretaceous (Santonian-?early Campanian) megaspore Glomerisporites pupus and its associated microspores has been examined in an attempt to resolve a number of problems concerning the interpretation of their morphology. The new observations presented are based on an analysis of entire, fragmentary, and thin-sectioned specimens under scanning and transmission electron microscopes. These add to, and partly correct, previous observations on this taxon. They include the following: (1) The exine of the megaspore consists of thin, homogeneous, outer undulating and inner electron dense layers, with a thicker zone of spongy structure in-between. (2) The perispore (or perine) of the megaspore comprises four layers, in order towards the exterior: loose filamentous, dense filamentous, vacuolate, and columnar. (3) This is completely enclosed by a thick mat of hairs, which appears to be attached to the underlying perisporal layers by means of connections with a few of the "spines" that originate from the dense filamentous zone, and with some elements of the columnar perine, (4) The tripartite neck (acrolamella) of the spore, which is hidden beneath the mat of hairs, is predominantly an extension of the dense filamentous and vacuolate layers, but also involves the columnar layer, especially in the lower part. (5) Some of the numerous small floats that are embedded in the mat have hairs originating from them. (6) Both long tangled and circinate hairs surround the perispore of the microspores. (7) The exine of the microspore was at least partly attached to the perispore when the organ was viable. (8) It comprises four zones that vary in structure and electron density. These facts and comparisons made with other megaspores and their associated microspores confirm evolutionary links between G. pupus and several taxa included within the Salviniaceae (Azolla, Parazolla, Salvinia) and possible ancestors of this group (the parent plants of Ariadnaesporites and Capulisporites).