160 The Phenomena of Morphogenesis 



been able to modify phyllotaxy operatively in a number of ways. Thus 

 in Epilobium hirsutum, a species in which the leaf arrangement is 

 decussate (opposite), the Snows (1935) split the apex diagonally and 

 found that the two regenerating shoots had spiral phyllotaxy. They were 

 also able ( 1937 ) to change the phyllotaxy in the same way by applying 

 auxin to the shoot apex. There may be a rather delicate balance between 

 decussate and spiral phyllotaxy in this plant, for in the group to which 

 it belongs (and even in a single plant of this species) both types may 

 occur. In other plants the direction of the phyllotactic spiral may be 

 reversed in regenerating shoots after splitting the apex. 



What determines the location of a given primordium is the basic prob- 

 lem here, and as to this there are two major hypotheses. One, first pro- 

 posed by Schoute (1913), assumes that the presence of a primordium 

 tends to inhibit the development of another one near it, presumably by 

 the sort of inhibition by which one bud checks the growth of another 

 through the agency of auxin. This is the same problem studied more 

 recently by Biinning (p. 199), who has evidence that each stoma produces 

 a substance that prevents the development of another stoma close to it, 

 thus accounting for the regular spacing of these structures. Such a hy- 

 pothesis is in harmony with physiological theory, but some experimental 

 results seem to be opposed to it. For example, the Snows (1952) re- 

 moved the youngest actual primordium in an apex of Lupinus and after 

 14 days determined the positions of the next three successive primordia 

 that had appeared since this was done. In every case these later ones 

 occupied the places in which they normally would have appeared, in- 

 dicating that their positions in the series had not been affected by removal 

 of a primordium and any inhibitory influence from it. 



That the primordia develop independently of either stimulatory or in- 

 hibitory influences from neighboring ones is also shown by an experi- 

 ment of Wardlaw's in which he isolated by radial cuts the areas pre- 

 sumably to be occupied by the next primordia in the series, thus effec- 

 tively isolating them from physiological contact with primordia already 

 formed. He found that these areas developed primordia normally. 



Other factors than chemical ones may here be involved. Wardlaw 

 ( 1948 ) finds that each primordium tends to produce a region of tangential 

 tensile stress around it but that this is absent in the area where new 

 primordia are to arise. He suggests that a primordium will develop where 

 tensile stress is at a minimum. 



The second hypothesis assumes that a primordium will not develop 

 unless there is sufficient available free space for it. This is related to the 

 ideas of Schwendener and Hofmeister and really comes down to the 

 problem of the most efficient filling of the space on the surface of the 

 meristem. It has been supported, in essence, by van Iterson (1907), and 



