IMAGE5.jpg (53785 bytes)


From the Red River Floodway
Winnipeg, Manitoba, Canada

By Ron Zeilstra

"Rosettes" of gypsum crystals have been found in the banks of the Red River Floodway in Winnipeg, Manitoba since the late 1960’s. Greg Hasler and I spent 2 weeks in May 1990 prospecting for productive zones and excavating one rich pod of crystals.

Image2.jpg (38529 bytes) The Red River Floodway is a 49-kilometer man-made diversion channel running along the eastern boundary of the city of Winnipeg.  It was built to reduce risk of spring flooding in Winnipeg by diverting a portion of the water from the Red River to outside the city limits. It took 6 years to construct and was completed in 1968 at a cost of $65 million. The floodway has an average depth of 10 meters, and the width ranges from 130 to 180 meters at the base to 230 to 330 meters at the top of its banks.

The floodway cuts into a flat clay plain which expands from south of Lake Winnipeg into North Dakota and Minnesota. The plain was created in the Pleistocene when glacial Lake Agassiz covered most of southern Manitoba. Typically, the clay is overlain by up to 10 feet of sediment deposited by the regular flooding of the rivers in the Red River Valley.

IMAGE3.jpg (84219 bytes) The rosettes occur in the glacial clay layer. Gypsum in the clay may have formed from the retreat of Lake Agassiz or may have originated from underground Jurassic gypsum deposits located just west of Winnipeg. The rosettes are not found uniformly distributed throughout the banks of the floodway but occur in small "pods" of high concentration in certain areas of the floodway. Within the pods there appears to be horizontal banding of rosette forms, which may reflect changes in the permeability of the clays as well as the effect of depth and unfavorable growth conditions nearer to the surface.

Rosette pods have been reported from many locations in the southern segment of the floodway. The crystals are not restricted to the floodway, but it is one of the few structures that cut deep enough into the floodplain to intersect the clay zones. Anecdotal evidence suggests that occasionally crystals have been found when digging the foundations of large buildings in Winnipeg. Rosettes have also been found in excavations near Birds Hill Provincial Park, 15 km northeast of Winnipeg, and in the banks of the Red River near the floodway gates.

There are two typical forms of rosettes found. In shallower depths, the standard is an amber colored compact ball of intergrown crystals with small, thin blades pointing out from the core. Sometimes large transparent amber blades, often twinned, protrude out of this core producing spectacular specimens

IMAGE6.jpg (82500 bytes)

In deeper layers, the crystals in the rosette are larger, more distinct and blocky. The color in these specimens are typically yellow, but can also be colorless. Large blades protruding from these rosettes are also blocky. In both forms, some of the large blades may have clay or a rock included. All the crystals are fluorescent and phosphorescent, glowing a pale white under ultraviolet light.

I had collected the rosettes as a teenager with the local rock club and my partner had dug crystals in the Floodway in previous years. Greg believed the pod near the Trans Canada Highway, which he had worked the previous summer, was practically exhausted and we decided to do some prospecting for a new site. After talking to several seasoned floodway collectors, we had a list of areas with known pod occurrences and some tips on how to recognize surface features, such as vegetation and clay coloration differences, which could indicate the presence of crystals.

IMAGE8.jpg (78602 bytes) IMAGE10.jpg (90348 bytes)

We spent a week of prospecting and digging holes with little success. Usually the pit had to be 4 to 5 feet deep before we knew if there would be any crystals. Winnipeg’s harsh winters would destroy any crystals near the surface. We did find rosettes at one site, however the ground was very dry and the rosettes were unextractable. We abandoned our excavation there and continued our prospecting. We did not find crystals at any of the other promising sites, even when we used the "surface indicators". Given the large areas, the small size of the pods, and our inexperience, this probably was not totally unexpected.

Not to be discouraged, we then decided to redirect our efforts on the proven zone off the Trans Canada Highway. We now know that there are several pods on both banks of the floodway in this region, but at that time we only knew of a site near the CN Rail trestle.   I couldn’t make it to the site that day, so Greg had a friend drop him off.   He hit a layer of frozen ground after digging a couple of feet. If he had brought a vehicle with him, Greg probably would have given up. But since he wasn’t going to be picked up for 8 hours, he decided to try to dig through it. The frozen layer turned out to be about 2 feet thick. After digging all day, Greg encountered unfrozen ground and a pod of crystals.

I was back the next day, and had to watch Greg pull out crystals while I struggled through the frozen ground. But there is much more incentive to keep digging knowing that there will be a reward for all your effort. By the end of the day, I was through the frozen zone and pulling out crystals as well.

During the next few days we expanded our pit in width and depth. We "scanned" the clay for crystals using long, straight wire probes. The traditional tool for doing this is a straightened coat hanger. The probes are pushed gently through the clay until they meet with resistance, usually indicating a crystal. Often there were groups of crystals and the surrounding clay had to be probed to determine a safe path to excavate.

Once we got close to a rosette, we would notice a difference in the clay. It would peel away in a pattern similar to a concoidal fracture and would have visible moisture on the peeled faces. We would dig around the crystal leaving a protective layer of clay about a centimeter thick. A garden trowel would be used to do the fine digging, with more probing to determine if there were any protruding blades on the rosette.  We could not pry out the block containing the rosette because the resulting pressure would destroy the crystal. Even with taking extreme care, we lost many crystals.

By the end of each day we looked like we belonged in a Tide commercial, covered with drying clay. The clay balls containing the crystals would be protectively wrapped with newspaper and taken back each night for cleaning. Usually a short soaking in luke-warm water would loosen the clay which then could be peeled away. Then a low force stream from the faucet and a toothpick would clean up the rest. If the crystal consisted of blocky blades, too much water could dissolve the cement binding the blades together and the crystal would fall apart.

Greg and I spent about a week collecting from this pit. It was eventually 9 feet deep with the potential to produce rosettes even deeper. Eventually we ran out of time and the pit was filled in.  I went back to the site a few years ago and all I could see was a bare patch where the vegetation is just starting to grow and an indentation in the ground where the pit used to be.

Image15.jpg (120220 bytes) Greg still collects gypsum crystals in the floodway. When I visited his new site two year ago, an associate was extracting crystals from a pit 16 feet deep.

rrfw16.jpg (55580 bytes) Last summer Greg and his associates brought in a backhoe to this new site and had a major excavation going. It will be interesting to see what they found.

Table of Contents

Bob Keller Copyright 1998 Ron Zeilstra