This caddis fly is a member of the Family Brachycentridae
(the tube-case makers). Here the fly I fish is Brachycentrus
americanus. This is just one of nine members of the genus.
All members of this genus make cases. This larvae makes a four-sided
black case, about 15mm long, that appears to consist of wood, which
it holds together with an interior silk lining. The larvae itself has a
light green body with dark legs and a real dark head. The body, head,
and legs just peak out of the case when its active and at the first sign
of danger it pulls all of its parts back in like a turtle. It feeds by scraping
algae off the rocks by day and using its legs like a net to trap food during
the night. The unique thing about the larvae is that they live in the rocky
riffles and anchor themselves to the rocks by a short white silk line. It
uses that line to rappel from stone to stone, the line serving to prevent
the case from being washed away.
The pupae appears brownish green
and pops to the surface fast enough that pursuing trout will often clear
the water in the heat of the chase. The pupae may drift for long distances
a foot or two under the surface before it pops to the top. The adult is
a brown-olive bodied fly which becomes really important during the
egg-laying process. I see the egg layers lying spent on the surface. I
believe some fail to break through the surface tension and drop their
eggs there while others that do succeed in breaking the surface swim
down to deposit their eggs on underwater objects, returning to lie spent
on the surface if they can break through the surface film. I have never
observed eggs on the spent flies. The spent flies I have observed have
laid there without movement. In other hatches the spent flies exhibited
a lot of movement.
What happened differently here? Did they all deposit their eggs under
water and the ones we observe on the surface are all returners that
never had the strength to fly off? Those that return to the surface and
fly off, which I believe some do, mate and return again to lay more
eggs. I believe this happens not only in this species but in many others
also, especially the Little Black Sedge, which interacts with this
species and the Hendrickson hatch in a complex manner. Little is
in the literature, and I have spent a lot of time searching. I attribute
the underwater activities to the fact that a sunken fly with a large
egg sac produces. The double egg-laying event or maybe triple is
an evolutionary positive thing I suspect. In the future, hundreds of
thousands of years or so, I think all or most of the surviving caddis
species will exhibit a multiple egg laying event.
I have had some limited success fishing the cased caddis in the
riffles weeks before the hatch with a ten to twelve inch drop of
the rod straight down stream. I haven't fished this way for years,
but I remember getting an opaque, white, water proof marker
from the local artist shop and marking about 20 inches of my
leader up from the fly. I only used it when nothing else was
happening, and during this time of the year that was a rare event.
I discontinued the practice when I came to the conclusion that
trout were taking the fly not as a larvae, but as a something that
might be good to eat. A poor attractor and a stupid choice. The
trout caught on cased imitations can attribute their demise to the
buggy look of the fly. Paul Young's strawman nymph is a nice
impressionistic fly in its own right. To give the devil his due I
have seen plenty of these cases in trout stomachs.
There is a complex interaction between the Little Black Sedge, a
size 18, and the Hendrickson and the Grannom which is a 16.
Early on in the Hendrickson hatch the Sedge precedes it, while later
the Grannom replaces the sedge. The only difference being the size
of the fly. I fish both dry caddis imitations almost always with an
egg sac as I believe the trout remember the egg laying event, into
the following day, which is the major event of the hatch. I have had
a little success with a spent Delta caddis but the fly I always return
to is the EHC size 16 which has a yellow egg sac on the terminal end.
We are all in the same pitch-pot. At this time of year there can be six
or so flies the trout are on, and wise people don't preclude the Grannom.
I fish the Grannom and the Little Black Sedge as they compliment the
Hendrickson hatch, like I should. In the early morning and afternoon
its caddis and later it's Hendricksons. I use a pupae imitation big time
during this hatch. Some accuse me of using a small split shot twelve
inches above the pupae under an indicator on bad days.
The real producer is usually a pupae imitation fished down and across,
most of the time dragging on the surface, fished with a rhythmic twitch.
Al Campbell did you patent this presentation?
When egg laying begins, a diving caddis with a pronounced egg sac
might be the thing. Fish it weighted with a series of short pulls, down
and across. An EHC that has a wing that doesn't float it in a hurricane
should work.
Here in the east be ready to switch to the Hendricksons when they
show up in the mid-afternoon.
This is not a stand alone hatch, as most probably are not, but it
should be a major component of most spring fisher's act.
This time of the year you can find trout on one stage or another of a
group of flies. There can be the leavings of the Tiny Winter Blacks,
Baetis and pseudocloeon will be there,
the Sedge and the Grannon will be in forceful attendance along with
pockets of several others.
The important thing to remember is that the same fly will come
off the same part of the stream every year. Grannoms will be
Grannoms, and they always seem to fit themselves into natures
march at the same place.
Here's what I don't know:
1. Do most dive below the surface to lay their eggs? I think they do
but I can't find real observations of that event.
2. Do high concentrations of the Little Black Sedge preclude the
presence of Grannoms, and visa versa? I see Sedge riffles and I
see Grannom riffles. Why? They look the same to me.
3. Does the pupae really drift as far as the literature says it does?
Most of my fish come from a dragging fly on the surface. I suspect
the long drift may be an entomologist's dream.
4. A high percentage of my fish are accounted for by a dry EHC. The
literature say that should not be. Is it just because I fish the dry more?
I don't think so. There is a big point missing here! I tie the body with
peacock herl, or if not then I tie it black as coal.
I don't think the tie is the important thing. I believe its accessing the
correct stage of the insect at the right time.
There is a right girl to take to the dance. Old Rupe
Special credit to: Aquatic Entomology by W. Patrick
McCafferty, published by Jones and Bartlett for the technical illustrations in this
article. Other photos from Penn's Creek River Journal
Frank Amato
Publications, Inc.
|