Electronic Effect of Suhstituents on Phenoxyacetic Acids 445 



trogen indene-3-acetic acid happens to be the first analogue of indole- 

 acetic acid to have been studied. If we look at these four substances, 

 the molecules are the same in general shape. Originally it was thought 

 that they are all active because they all have the same shape. It is 

 now clear that molecular shape is not nearly as simple a criterion as 

 it sounds, for there are compounds like 2,3,6-trichlorobenzoic acid 

 u'hich are highly active, and of greatly differing shape from that of 

 lAA. So we have to ask, What do these molecules have in common 

 other than their shape? One answer is that there is in all four of 

 them a strong fractional positive charge on the atom at the bottom of 

 the 5-membered ring. In the case of nitrogen there is a lone pair of 

 electrons which is drawn into the ring, leaving the nitrogen posi- 

 tively charged; oxygen and sulfur have similar lone pairs that are 

 drawn in and are left positive. Furthermore, there is good evidence 

 that cyclopentadiene and indene have exactly the same property on 

 this methylene atom as the other cycles have on the hetero atom. 

 The cyclopentadiene or indene ring is sensitive to acids; the hydrogen 

 is replaceable by potassium and it combines with NO. Furthermore, 

 the dipole moments of all these rings indicate a marked fractional 

 positive charge. The first thing that is common to these four mole- 

 cules, then, is the fractional positive charge in a characteristic position. 

 Now, let us look at the other end of the molecule. What is char- 

 acteristic of all auxins is the carboxyl group, which can ionize, so that 

 here we have in a characteristic position a potential negative charge. 

 Now, of course, the idea is current that the carboxyl reacts chemi- 

 cally to form an amide type linkage or an acyl group of some sort, 

 but I feel that the evidence points against that. Perhaps the strong- 

 est reason for thinking so is, as Dr. van Overbeek has pointed out, 

 Veldstra's work with the tetrazoles. It is almost impossible to see 

 how a tetrazole could form any kind of an acyl type linkage, and 

 yet the one thing that the tetrazole does do is to dissociate an H+ and 

 thus produce a charge at this point. I suggest that there may be no 

 true chemical reaction in the sense of covalent bonds, and that we 

 have to deal rather with the approach to a receptor, based upon the 

 electronic configuration of the molecule. The distance between these 

 two charges, of course, varies — there is free rotation — but it centers 

 around 5i/2 angstroms. Dr. Wain has presented evidence that long 

 side chains are reactive only after beta oxidation. An exception is 

 made in the case of propionic side chains which are apparently ac- 

 tive in themselves. Correspondingly, in p-chlorophenoxypropionic 

 acid, for instance, the distance between the positive charge at the 

 ortho position and the carboxyl is not much greater than the desired 

 order of magnitude. In p-chlorophenoxyacetic acid it is just right. 



