360 THE JOURNAL OF BOTANY 



A. B. C. D. 



9. 



40 



28 



•20 



40 



10. 



40 



27 



20 



41 



11. 



40 



25 



18 



42 



12. 



39 



26 



18 



38 



13. 



38 



34 



19 



88 



14. 



35 



20 



14 



37 



15. 



33 



25 



18 



36 



10. 



32 



11 



15 



32 



17. 



31 



31 



17 



37 



18. 



31 



25 



15 



38 



19. 



31 



19 



16 - 



36 



20. 



31 



16 



16 



40 



21. 



30 



19 



14 



29 



22. 



30 



15 



15 



38 



23. 



30 







15 



30 



24. 



28 



20 



15 



30 



25. 



28 



17 



15 



38 



20. 



28 



6 



12 



33 



27. 



20 



12 



15 



45 



28. 



26 



10 



10 



33 



29.* 



25 



6 



12 



41 



30. 



25 



5 



12 



31 



31. 



25 







10 



28 



32. 



23 



15 



12 



28 



33.* 



23 



12 



12 



38 



34. 



23 



9 



12 



33 



35. 



23 



6 



13 



31 



30. 



22 



17 



11 



31 



37. 



22 



8 



11 



37 



38. 



22 



7 



11 



29 



39. 



21 



4 



10 



35 



40. 



21 



4 



10 



32 



41. 



20 



8 



11 



31 



42 .-•'■ 



20 



4 



10 



32 



It will be observed from this table that, as in the CE. biennis 

 tabulated above, the range of variation in the length of the tube 

 and of the stamens is not nearly so great as that in the length of 

 the petals and of the pistil." De Vries himself has remarked that 

 the flowers of Gfi. muricata, which are not half so large as in 

 CE. biennis, are borne by a calyx-tube of the same length. 



Leaving out two aberrant specimens (Nos. 23 and 31), in which 

 the style does not project beyond the tube (as in the CE. brevistyUs 

 of De Vries), we see that the length of the style (outside the tube) 

 varies from 4 to 37 millim., and the length of the filament from 

 10 to 22 millim. In the large flowers (petals 33 to 48 millim.), 

 the pistil is always much longer than the stamens, as in the 

 CE. Lamarckiana of De Vries,! the variation in the length of the 



* Range of variation as observed at : 



A. B. C. D. 



South Kensington 10-24 8-18 7-13 17-27 



St, Cast.. .. 20-4S 0-37 10-22 28-45 



t The following are the measurements of two flowers of the typical 



