140 



PYRENOMYCETES [CH. 



The Hypocreales, Dothideales and Sphaeriales, have in common more 

 or less pyriform or flask-shaped perithecia ; these are sometimes isolated and 

 free, sometimes sunk in the tissue of the host, and sometimes embedded in 

 a stroma or cushion of fungal tissue. The perithecium is lined by delicate 

 filaments, some of which, the periphyses, grow along and partially close 

 the neck, and may protrude through the ostiole, while others (paraphyses) 

 are mingled with the asci in the venter of the fruit. The neck of the peri- 

 thecium varies very much in length, and is often markedly phototropic, the 

 ostiole being directed towards the light, and thus incidentally towards a clear 

 space so that, when the spores are shed, as wide a distribution as possible 

 is ensured. So definite is this reaction in, for example, species of Sordaria, 

 that if the direction of light be changed every four or five days during 

 development, a series of corresponding bends in the neck are produced. In 

 Sordaria and its allies the asci elongate, reaching up to the ostiole and in 

 turn discharging their spores; in species of Spumatoria and Ckaetomium the 

 asci deliquesce to form a mucilaginous mass which readily absorbs water 

 and expands, being squeezed up the neck and exuded at the ostiole where 

 it persists until dissolved by rain or dew. 



Accessory fructifications include chlamydospores and various types of 

 conidia which may be borne separately on free conidiophores, or grouped 

 together in pycnidia. In some cases there is evidence that the so-called 

 pycnidia are spermogonia, and the spores they produce spermatia, but no 

 case has been brought to light in which these still fulfil their function as 

 fertilizing agents. 



A consideration of our rather scanty knowledge of the initiation of the 

 perithecium in this group brings to light three main types of development, 

 (i) In Chaetomium, in Sordaria (fig. 101) and its allies and in species of 

 Hypomyces and Melanospora there is a coiled archicarp of four or five cells; 

 these are uninucleate in Hypomyces lateritus, 

 Chaetomium spirale and Podospora hirsuta, 

 multinucleate in Sordaria and Hypocopra and 

 in other species of Chaetomium. 



In Sordaria macrospora the archicarp is 

 straight instead of coiled and in 5. fimiseda 

 a swollen terminal cell has been reported. 

 A pair of initial hyphae has been described 

 in Rosellina quercina, but in no case has a 

 sufficiently detailed study been made either 

 to reveal nuclear fusions in the archicarp or 

 to justify the inference that they do not occur. 

 F" der these circumstances it is possible to 

 judge of the function of these initial filaments 



