50 ON A GENERAL METHOD OF SUBSTITUTING 
atom of chlorine simply displaces or pushes out the atom of hydrogen, and so 
comes to occupy its place, but a very slight consideration enables us to see that 
it is really dependent, not so much on the quantity of chlorine which remains in 
the compound as on that which escapes in combination with hydrogen. The 
original substance in all cases of substitution, forms a perfect molecular group in 
which the individual affinities of the different elementary atoms are properly 
balanced, and the whole remains in a quiescent state. But when another ele- 
ment, such as chlorine, comes in contact with this complex group of atoms, it im- 
mediately exerts its affinity for the hydrogen, with which it easily combines, and 
withdraws the whole or part of it from the compound. In this way a gap is 
produced in what was before a perfect group, the balance of the affinities of its 
elements is destroyed, and there must either be a complete readjustment of its 
molecular arrangements, or some other element must stop the gap and produce 
another perfect molecule not differing from the original substance in the arrange- 
ment of its parts, but only in the presence of one or more atoms of a different 
sort from those which it previously contained. Considered in this point of view, 
it becomes at once obvious that we fail to produce the substitution of iodine for 
hydrogen, not from any inability of the former to occupy the place of the latter, 
but simply because it has not a sufficiently powerful affinity for hydrogen to 
withdraw it from the compound, so as to leave an empty space into which an- 
other portion may enter. It occurred to me that as chlorine and bromine in 
causing a substitution exercise the two different functions already alluded to, one 
quantity withdrawing hydrogen, and the other simply slipping into the vacant 
space, it might be possible to produce an iodine substitution by associating that 
element with some substance having a sufficiently powerful affinity for hydrogen, 
to open the door as it were, and leave nothing for it to do but to step into the 
place prepared for it. 
For this purpose I selected the bromide of iodine, as being most likely to fulfil 
the required conditions. It was prepared by agitating bromine water with a 
considerable excess of iodine, and decanting the reddish-brown solution from the 
undissolved residue. When this bromide was added to pyromeconic acid, a 
change rapidly occurred, the solution became colourless, and iodopyromeconic 
acid was produced. Having succeeded in this way, I then tried the chloride of 
iodine, and having found it to act equally well, I made use of it in all my subse- 
quent experiments. The chloride of iodine was prepared by passing a rapid cur- 
rent of chlorine through finely-pounded iodine suspended in a small quantity of 
water, and kept in continual agitation, care being taken to stop the process before 
the iodine was entirely dissolved. A moderate heat is produced during the combi- 
nation, and the fluid should be kept as cool as possible. 
In order, therefore, to obtain iodopyromeconic acid, a freshly-prepared solu- 
tion of the chloride of iodine is mixed with a cold saturated solution of pyrome- 
