INTRODUCTION 373 



verse wall. At this stage the terminal cell now has two nuclei, the cell 

 below it has one nucleus, while the fourth nucleus lies in the lateral pocket. 

 The latter curves around until it is in contact with the lateral wall of the 

 upper end of the penultimate cell and the intervening walls are dissolved 

 and the nucleus passes into the latter cell. The lateral pocket has acted 

 as a by-pass through which a nucleus has been transferred from the 

 terminal to the penultimate cell in such a manner as to provide each cell 

 with a daughter nucleus of each of the two nuclei originally in the terminal 

 cell. This by-pass is known as a clamp connection. Buller (1933) has 

 followed the formation of these clamp connections in living mycelium and 

 found that the process requires only a short time. In Coprinus lagopus Fr. 

 the time elapsed from the first appearance of the projecting lateral pocket 

 until the passage of the nucleus out of the pocket into the penultimate 

 cell was 23 minutes, while in C. sterquilinus Fr. it was 40 to 45 minutes. 

 The conjugate division of the nuclei was completed in the first species in 

 from 12 to 14 minutes. (Fig. 124.) 



The phenomena involved in the formation of the clamp connections 

 are generally considered (e.g., by Kniep, Bensaude, and most students 

 since then) as being homologous to those occurring in an ascogenous 

 hypha when an ascus is forming by the hook or crozier method. The 

 Moreaus (1928) have shown that the formation of the hook does not 

 necessarily lead immediately to the formation of an ascus, for the terminal 

 binucleate cell may elongate and again form a hook while the tip of the 

 original hook unites with the cell below. This may continue several times 

 until a series of dicaryon cells is produced, each connected to the cell 

 below by a clamp connection. De Ferry de la Bellone (1886) and Mat- 

 tirolo (1887) described and figured typical clamp connections on the 

 mycelium of Tuber layideum Matt., and other species. Unless they mis- 

 took some intermingled strands of Basidiomycetous mycelium for that of 

 the fungi they were studying the occurrence of this structure in the 

 Ascomyceteae as well as in the Basidiomyceteae must be considered 

 substantiated. 



Buller (1933) did not beheve that the formation of clamp connections 

 is at all homologous to the processes occurring in the ascogenous hyphae. 

 The clamp connections play, he beheved, an important part in the trans- 

 fer of food through the mycelium. An actual flow of protoplasm was 

 observed by him through the clamp connection and its centrally perfor- 

 ated upper septum. Possibly the occurrence of whorls of clamp connec- 

 tions in some fungi would support Buller's view as to their function in 

 nutrition. Kemper (1937) studied the development and cytology of the 

 clamp connections which may occur in whorls instead of singly at the 

 septa of Coniophora cerebella Pers. (now called C. puteana (Schum. ex Fr.) 

 Karst.). In this species the uninucleate young basidiospore becomes bi- 



