Dear Bruce:

Got yours of the 3d. I want to reassure you that I'm still neck-deap
with the b'y trails; in fact I've been able to do nothing else since the
start of the year. After some fumbling again, I decided to concentrate
on SW-534 —x 43, which does give a remarkably high yield of profuse
trails. AnotWef/advantage is the rarity of swarms, so one can safely
economize on plates when plahting single cells.

I think mmm one just has to look closely at these trails to see
that they are not unilineas, but they do give just the impression you
suggest of one main track and numerous side branches. I have also had a few
not very impressive successes in getting two or more trails from one clone;
unimpressive, because most of these (side—)tracks are very weak. While this
may weaken the accuracy, or at least the generality, of the idea that only
pluri-catenate cells can form trails, they do not touch closely on the main
point: is there a single primary branch in each catenate system. I have been
hoping to be able to get at this by trail observations, planting sibs, atc.,
so far with rather poor look owing to inconsistencies in indidence of trails.
But yesterday, e.g., I picked out 894 (sic) motile initials, diluted them
in about .2 ml broth and plated this out in various ways. Pour plates look
the most promising; the incidence of trails is about 1/8 (which is mich higher
than the somtk incidence of strikingly pluricatenate clones, which is about
1/25) and theke appearance under these conditionsnot at all unilinear. I have
had indifferent luck imposing a definite chemotactiw gradient, viz. with phencl
or excess non-motiles, but shake tubes now may help. In the deep pour plates,
there presumably will be no significant metabolite gradient, and the aerobic
one should be minimized. One of these deep trails, by the way, counted to 125
wolonies at 17 hours, with a nice size gradfation of microcolonies to orient
with respect to time. The distribution of smaller colonies made it quite clear,
beyond the general appearance of these "clusters", that numerous side—trails
are indeed formed; I think they are minimized under surface plating conditions
because all the motile cells will be moving in response to a common gradient.
But even ghese clusters do seem to show a "unilinear" trend; it will be devil of
a job to pin this down a8 svidence of two hierarchies of unicatenation.

The trick in making so many isolations easy is just to concentrate the
treated bacterial suspensions about 10x before seéting up the trap drops. The
motiles then come out in dréves, partly, I think, from the tactic pressure as
well as from there simply being more of them.

A few (incomplete) comments on yours. At (11), e.g., I think the problem is
recognizing twin-trails under tactic orientation; the deep pours should help. I
had misunderstood you at (12). I do have some doubts, by the way, about the
randomness of distribution of few motile cells; plating out gives a direct
count. Spreading on surfaee is essentially as efficient. 14): if you
pat a drop of motile Salmonella under o11 about .3 mm. from a drop of
-5% phenol and watch 20-40 mins: see what happens. (I picked this up ac-
cidentally while preparing some tests with mbtility in serum: the serum
had some cresol preservative).

As to your¥ PPS, I am only a little worried about the specificity of

the srum inhibition of trails. I did not get it with wméaiena 1/ 1565 ;
cr some similar thing. It should be shown that ‘oxcmmemmortkkk i or b w
inhibit mearm trails from i--x SW543, but not, say, a—x SW~-543, and

vice versa, am I have not done this. I don't think this necessarily
proves that the and Fla, factors continue to be associated on a single
fragment, only + the initial cell does receive both. This would be

worth pursuing, however, as a possible distinction (hypothetically) between
pluricatenate "F" cells, and unicatenates. Indeed if unicatenates of SW553
are more vigorously motile, this might explain whey they were less affec-—
 

More. mich more will follow.

°

8 , o

Q

a

eo wee
Bw. OF ee oo
Sc ot ™ *
oo]

¢

3 .

~

3 2