THE BOUQUET STAGE 



91 



(e.g., Aphis, V. Baehr, 1912). Certainly the failure to observe it in 

 unfavourable material, such as many dicotyledons and Diptera, is 

 inconclusive, but there can be no doubt of the absence of 

 polarisation in many Lepidoptera (Seller, 1914), and in some Liliaceae 

 except merely as a relic of the orientation of the chromosomes, 

 with their centromeres towards the pole, at the preceding telophase 

 (Newton, 1927 ; D., 1929 h, 1935 a). Polarisation may be regarded 



Fig. 22. — Complete early pachytene nucleus in Cliorthippiis parallelus. 

 Eight paired chromosomes shown as single lines, three long 

 pairs broken or wavy. X chromosome ovoid. All ends 

 but two polarised. Small nucleolus in outline. X 3,000 

 (D., 1936^). 



as an adaptation to secure regularity in pairing [cf. Gelei). It is 

 bound to affect the distribution of pairing amongst the possible 

 partners, where there is a choice, whether amongst segments in 

 structural hybrids or whole chromosomes in polyploids. 



With a bouquet arrangement pairing begins at the end or ends 

 lying towards the surface of the nucleus, whether the centromere 

 is situated there or not (Fig. 22 ; Wenrich, 1916 ; Gelei, 1921 ; 

 Belar, 1928 b). In the absence of a bouquet it is more difficult to 

 trace the course of pairing except where the chromosomes are 



