our govt. says there are no long term studies on Marijuana

torontomeds

Well-Known Member
The University of Guelph has been growing MJ in a guarded section in one of their greenhouse on Gordon St. since the early 90s. Seen it with my own eyes making a delivery of Fert's....never heard why?
Is it possible that you saw Hemp? I know in the 90's they had been working on Hemp strains for the "Thunder Bay Hemp Growers’ Association" could it have been some hybrid hemp in flower? I asked a few of my buddies who work in the secure part of the Ag dept and they said the only Cannabis plants are just now arriving via ABcan and Conviron.


ABcann and the University of Guelph are collaborating in the design of closed environmental grow systems

The University of Guelph is ranked as one of Canada’s top comprehensive universities because of their commitment to student learning and innovative research. They are dedicated to cultivating the essentials for our quality of life – water, food, environment, animal and human health, community, commerce, culture and learning. The University community also shares a profound sense of social responsibility, an obligation to address global issues and a concern for international development.

Guelph is the top-ranked university in horticultural education in Canada and through partnered research at the Controlled Environment Systems Research Facility Guelph scientists have created the most advanced technology globally for sealed environment plant growth production systems. At Guelph, CESRF researchers work with international space agencies (including the Canadian Space Agency) in the development of long-term life support requirements (provision of food, water, recycling of CO2 etc.) for long duration missions and eventual habitation. ABcann and the UG-CESRF are partnering in research that addresses the specific growth requirements for the high fidelity standardized production of medical cannabis using customized plant growth chambers and highly customized LED lighting sources.

Established in 1964, Conviron is a world leader in the design, manufacture and installation of controlled environment systems. We offer a broad spectrum of products and services uniquely tailored to each client project. Our product offering ranges from standard products, to highly configured products, to custom solutions.

As the world’s largest supplier of plant growth chambers and rooms and high performance research greenhouses, our primary markets include agricultural biotechnology and other life-science disciplines. A growing portion of our business is the industrial market sector where we offer custom-engineered environmental rooms (E-Rooms) for a variety of applications such as material testing, environmental simulation, stability testing, high-value storage, biosafety level containment, and other specialty applications.

I have been in the area with all the indoor grow chambers at UofG, the way it works is the University is partnered with all kinds of industry so some of the chambers have nothing to do with the UofG other then it is housed there and the students sometimes work in them, most of them are operated by Pharma, Science and Ag companies, Monsanto is there, NASA is there, DARPA is there and a few other companies you would not have heard of but they are doing some crazy things. One of the companies is taking antibiotics proteins and splicing them into tobacco plants to harvest cancer fighting drugs, I was in one of the labs and got to watch them do the TC and inject in the genes. It is a really cool place doing some super cool things, sure some of the anti GMO people would not like it, but hey they are making drugs to help fight cancer.

Our patented molecular farming platform, vivoXPRESS®, uses tobacco plants to produce high quality monoclonal antibodies and therapeutic proteins for the treatment of cancer, HIV/AIDS and other devastating diseases. With production costs approximately one-tenth of the cost of mammalian-cell systems, vivoXPRESS® is our solution to the rapidly rising global demand for more affordable biopharmaceuticals.

  • three biosimilar cancer drugs, leading with biosimilar Herceptin® (trastuzumab) for breast cancer
  • innovator antibodies for HIV/AIDS (R&D funded by the Gates Foundation and the Government of Canada)
  • innovator antibodies for treatment of Ebola virus (Sudan strain)
  • recombinant butyrylcholinesterase (BuChE) for protection against nerve agent exposure
  • three biosimilar cancer drugs, leading with biosimilar Herceptin® (trastuzumab) for breast cancer
  • innovator antibodies for HIV/AIDS (R&D funded by the Gates Foundation and the Government of Canada)
  • innovator antibodies for treatment of Ebola virus (Sudan strain)
  • recombinant butyrylcholinesterase (BuChE) for protection against nerve agent exposure
 

torontomeds

Well-Known Member
my Dad was in security back in the late 60's and had a gig guarding the govt. marijuana crop at the experimental farm in Ottawa if I remember right they had 4 acres surrounded by a tall fence with razor wire then another one about 50 ft. away they had dogs between the two fences and four watch towers manned with guards. no guns just billy clubs he said groups of hippies would take turns rushing it thinking they could not get all of them and the waves would go all night. did they grow this just to look at it or were they studying it? it is the experimental farm after all run by the Canadian Govt.
I seen some article about this place somewhere a few years ago, I wish I had it still, this place was real and they make all kinds of crazy strains, all the original seeds in the MMAR ad for PPS came from here. I heard some locked up forgot about cabinet in Ottawa is housing all kinds of landrace strains from this grow.
 

nsbudca

Well-Known Member
Never had the pleasure of trying a landrace sativa unfortunately, there aren't many around. Those were before my time.

I remember seeing seeds for a strain named after Pablo Escobar. Escobars gold maybe? I don't remember which breeder and I can't seem to find it anymore.

If I ever had the chance to grow I'd love to test it out, some landrace Colombian genetics. Anyone seen this?
 

torontomeds

Well-Known Member
Ok so I found the article and it was even better then I remembered. Here you go, all about the secret Canadian Gov grow. It was mainly a breeding project. Robert C. Clarke. interviews the lead guy in charge of the breeding project. He is still active too.
Enjoy.

Dr. Ernest Small is a Principal Research Scientist at the Eastern Cereal and Oilseed Research Centre, in Ottawa, a section of Agriculture & Agri-Food Canada (AAFC), the Canadian federal government ministry of agriculture. He is the author of seven books and 200 scientific papers dealing with economically important plants, including about two dozen publications on Cannabis. He has testified as an expert botanical witness in dozens of courts cases in North America, and has been a research consultant to governments in the United States and Canada on legislation concerned with plants, particularly Cannabis. Dr. Small has received several professional awards and honors for his scientific achievements. This e-mail interview was conducted in December 1999 by Robert C. Clarke.

What is your academic background? Have you ever taught at a university?

From 1963 to 1966 I obtained three degrees from Carleton University in Ottawa (B.A., B.Sc. (Hons.), M.Sc.), and in 1969 a Ph.D. from the University of California. Since then, I?ve been employed with AAFC. I have occasionally taught university courses, held adjunct professorships, and supervised theses, but most of my career has been spent as a research scientist, with occasional management responsibilities.

How did you come to work with Cannabis?

In 1970, our federal health ministry (Health Canada) became interested in growing a standard supply of medicinal marihuana for pharmacological research, and in clarifying some basic botanical issues about the species for forensic purposes. As well, the Le Dain Parliamentary Commission undertook a very extensive examination of the medical, social, and legal issues concerning Cannabis in Canada. AAFC placed me in charge of the botanical research, as well as the cultivation of the supply of medicinal marihuana. And, I was seconded to assist the Le Dain Commission half-time, by preparing a report on agricultural and botanical aspects of Cannabis.

Who directed that the Agriculture Canada Ottawa Cannabis farm be started and what was your initial mandate?

As noted above, cultivation of Cannabis was initiated by two Canadian federal ministries, Health Canada and AAFC, for the purpose of clarifying a number of botanical issues, and to grow a standard supply of medicinal marihuana. A 3 acre plot was established in 1971 in the main agricultural research area in Ottawa, known as the Central Experimental Farm, where both high-THC and low-THC forms were cultivated. This was in a rather open location, and elaborate security measures proved necessary. After 1971, it was decided to conduct cultivation in Ottawa only in more secure indoor areas, and this lasted until 1979. Outside of Ottawa, hemp has occasionally been grown outdoors at various AAFC stations across Canada in recent years, but currently cultivation is entirely on the private property of cooperating entrepreneurs. I am cooperating in research on hemp, both in the field and greenhouse, but the research sites are not in Ottawa. Our current mandate is to assist the private sector in conducting commercially oriented research on low-THC Cannabis, i.e. industrial hemp.

Did anyone else have a Canadian permit to carry out Cannabis research during this era?

Research cultivation of non-narcotic hemp for commercial development has been carried out under license in Canada only since 1994, and commercial cultivation only since 1998. In the preceding several decades, there was occasional growth of Cannabis under license for various pharmacological and forensic research purposes. Until 1994, virtually all of the authorized cultivation that took place in the last half century in Canada was oriented in some fashion to issues of law enforcement or pharmacological research, and not to the possibility of commercial cultivation in Canada.
 
Last edited:

torontomeds

Well-Known Member
Pt 2
Had you been following the international Cannabis research literature and attending conferences?

My first 10 years as a professional scientist were mostly devoted to Cannabis, but after that most of my research was on other crops of more immediate priority to Canada. Nevertheless, one doesn?t loose interest in a plant as amazing as Cannabis, and my literature collection on it has accumulated to occupy several filing cabinets. My return to working on hemp in the last 2 years reflects a widespread perception that hemp represents a promising new crop for Canada. I was a speaker at the international convention of the Hemp Industries Association, held at Hockley Valley, near Toronto this past September, attended the North American Industrial Hemp Council annual meeting in Chicago in November, and hope to attend upcoming major meetings.

How did you obtain seeds for the Ottawa Cannabis farm?

Curiously, it was easier to obtain seeds of Cannabis for research purposes in the early 1970s than it is today, and by 1971 I had accumulated in excess of 400 different stocks of seeds. The United States Department of Agriculture supplied dozens of samples, but today informs inquirers that it has none. Similarly it is also increasingly difficult today to obtain hemp seeds from some other seed banks. In my early research I concentrated on means of distinguishing drug and non-drug strains, and numerous police forces around the world cooperated in supplying seeds from confiscations. Today the emphasis is on non-drug strains, and it is necessary to have a reasonable indication that imported seeds are non-drug. And of course, with the recent realization that hemp is an extremely promising crop for which seed collections are critical for future breeding, many private seed collections are available only on a cooperative developmental basis.

Which accessions matured in the Ottawa climate?

Cannabis is a short-day plant, and both wild hemp and cultivars selected in various regions of the world are rather precisely photoperiodically attuned to flowering and maturing seed according to the latitude of origin. This means that southern European cultivars often mature seeds too late in the season when grown in Canada, which may not be a problem if one simply wishes to harvest fibre, but is a problem where the objective is to harvest the seed. This is a critical issue for Canada since most hemp produced in this country is dual-purpose, i.e. harvested for both fibre and seeds. High-THC forms that originate from southern Asia and Africa also are adapted to much longer growing seasons than found in much of Canada, and usually merely start to mature seeds before being killed by frost in Canada. This is advantageous in at least one respect: pollen from illicit marihuana plantations is less likely to contaminate the seed produced in fields of industrial hemp. Although the suggestion has been made to grow more than one crop in a season, this doesn?t seem practical in the short season of Canada. We have grown extremely early-maturing cultivars, but these matured in midsummer, too early to produce much of a harvest. It would seem preferable to have cultivars that take advantage of the entire photosynthetic season, maturing in the fall. The majority of the more than 60 accessions we grew in our Toronto area plantation this past summer proved to be suitable in maturation time. However, several of the more impressively yielding accessions were too late-maturing, pointing out the need to breed cultivars suited to Canada.

What did you learn from growing so many varieties of industrial hemp and drug Cannabis? What was the impact of your research at the Ottawa farm?

As a result of our botanical and chemical studies, we established a formal botanical classification system for Cannabis, that has been very widely adopted. We recognize only one species - C. sativa L., and as described in my 2-volume book, The Species Problem in Cannabis, Science and Semantics, this was critical to settling a forensic debate over the existence of "legal species" of marihuana. Drug forms were placed in subspecies indica while non-drug forms were placed in subspecies sativa. Within subspecies indica, all domesticated forms were placed in one variety, var. indica, and the wild plants were put in a second variety, kafiristanica; and similarly within subspecies sativa the domesticate was recognized as var. sativa and the wild phase as var. spontanea. Since the cultigens intergrade with each other and with widespread weedy forms, all classifications of C. sativa are necessarily inexact. One element of this classification has proven to have a huge impact - the way we separated the "non-drug" subspecies sativa from the "drug" subspecies indica on the basis of a dividing line of 0.3% THC, dry weight basis, in the leaves and flowering parts of the plant. This level of 0.3% THC is now used in the European Economic Community and Canada as a criterion for authorization of cultivation. Cultivars with less than 0.3% may be legally cultivated under license, whereas those with levels of 0.3% or greater may not. In the U.S., the figure of 0.3% has been suggested as a basis for allowing importation of hemp materials for non-human food use.

What part did the United Nations play in your research?

In the 1970s, the United Nations subsidized some of my publications and travel to international conferences. During this period the UN was instrumental in facilitating contacts among scientists working on drug control aspects, and the backing of this prestigious organization facilitated obtaining seeds and information. Of course, the major objective back then was controlling the drug problem associated with Cannabis, not developing industrial hemp as a crop.

Did you have a relationship with Dr. Carlton Turner?s research group at the University of Mississippi?

Indeed we had cordial relationships with Dr. Turner and the Mississippi group for many years, exchanging information and keeping each other abreast of developments. As you know, after his work on Cannabis Dr. Turner went on to occupy the prestigious position of Domestic Policy Advisor to the president of the US. We used a Mexican strain supplied by Dr. Turner to grow a large supply of standardized medicinal marihuana in Ottawa for Canadian use. Climatic conditions in Mississippi and Canada are of course different - a longer, hotter, drier situation in Mississippi, and this resulted in the plants in Mississippi tending to be taller, with much more leaf fall, leaving the remaining younger leaves with higher THC content on the plants. Nevertheless, morphological and chemical characteristics tended to be comparable between the two locations.

What are your current responsibilities at Agriculture Canada?

In addition to my work on hemp, I?m carrying out research on medicinal crops, vegetables and culinary herbs. I recently published three books on these topics, and am working on several additional books. There is a common theme to all of my current work - assisting the Canadian farmer to find new crops that will serve to reinvigorate agriculture. In this regard, I?m extremely enthusiastic about hemp, which has great potential to become a major crop in North America.
 

torontomeds

Well-Known Member
Pt 3
Do you currently work with international agencies or the United Nations? Are you collaborating with other Cannabis researchers?

The Canadian federal government has policies that encourage formal research collaboration agreements between government scientists and the private sector, and my research on Cannabis is mostly concerned with such collaboration. I?m in cooperation with Dave Marcus of Hemphasis Unlimited, a firm located in Toronto. We are assembling a collection of germplasm and evaluating the agronomic importance of our collection for the development of Canadian cultivars. This past summer we grew more than 60 accessions, and will be producing a report in due course. I also have ongoing collaboration with various universities. The paper in this issue of the JIHA, co-authored with Suzanne Montford, is a result of university collaboration. An extensive review of weedy hemp in Canada will shortly be available written with Professor Paul Cavers and his student Tessa Pocock-Whitney, at the University of Western Ontario. I am a plant taxonomist by profession, and great international efforts are underway by taxonomists to create computerized databases with key information on the world?s plant species. Since I have long specialized on the Cannabaceae, I am naturally responsible for this family in several of these international projects.

How did you first learn that Europe was looking into growing industrial hemp? Did you imagine that hemp would ever be grown in Canada?

Like most scientists, I keep abreast of the literature by means of profile searches, so that I have been aware of research trends with respect to hemp over the years. The resurgence of development of hemp in Europe was reflected by an increasing number of research studies, notably in the last 15 years, and especially in the last 5 years. I confess that I did not foresee the day when commercial hemp cultivation in North America would resume.

When was the 0.3% THC threshold for industrial hemp first suggested and from where was it derived? Did this standard replace a previous one?

Over the years, different researchers have suggested different levels of THC as means of recognizing groupings within Cannabis. It has long been appreciated that "drug" selections of Cannabis, with high levels of THC, originated from southern areas of Eurasia and from Africa; and in contrast selections of Cannabis used as sources of fibre and oilseed originated from northern Eurasia and had very limited drug potential. As documented in my publications in the 1970s, we picked the level of 0.3% as a formal means of separating the two classes of plants on the basis of numerical taxonomic methods. We grew several hundred different accessions, recognized the two groups on a number of physical and physiological characteristics independent of THC level, and the 0.3% THC level was calculated to be the concentration that naturally best separated the two groups. As a specialist in plant classification, it has been extremely gratifying to have so much of the world apply our simple classification in a practical way. But I would caution that 100% separation of groupings within species is rarely achievable, and this is the case for Cannabis sativa. Because of seasonal and diurnal fluctuations in THC content, a given plant could have more or less THC than 0.3% at different times. There are also minor environmental influences on THC content, and variations and errors in sampling method will also influence reported THC levels. In Canada common sense has been exercised by the authorities when levels slightly exceeding 0.3% have been found (by law, hemp fields must be eradicated if the level is higher than 0.3%). It should also be appreciated that the level of 0.3% is well under the concentration of 0.9% THC considered minimal for psychotropic effects by some authorities.

When did the Canadian government first investigate the feasibility of industrial hemp? Did you know about the promise of industrial hemp farming in Canada before 1994 when Joe Strobel and Geoff Kime grew the first Canadian crop? Were you called in to assess the Hempline application to grow hemp?

I have advised the Parliament of Canada, Health Canada, and our Justice Department for many years on legislation and research applications concerning drug plants, particularly Cannabis, and so am familiar with how hemp came to regain respectability in Canada. Until the Strobel and Kime application in 1993, the idea of reintroducing hemp as a Canadian crop was not taken seriously, and indeed in common with the US, our authorities considered this unwise in view of the problems posed by narcotic forms of Cannabis. Key to the change in climate was educating the Canadian authorities that the word "hemp" or the phrase "industrial hemp" should be reserved for the non-drug class of Cannabis, and was different from the drug class that produces marihuana. By the early 1990s, it was becoming obvious that the hemp industry was developing in a responsible way in Europe. Canadian farmers and entrepreneurs saw the possibility that hemp could be grown in Canada just as in Europe, with appropriate safeguards. With increasing demand by the private sector, and with support from key members of the federal government, as well as several provincial governments, the climate had changed sufficiently to enact regulations permitting the development of a commercial hemp industry.

Did you advise on the varieties to be approved for planting in Canada and suggest stipulations for harvest?

A committee was established to advise Health Canada on such specific matters. While I gave advice to members of the committee, I did not directly participate in the specific regulations that were enacted. I myself am subject to the regulations that govern research on hemp in Canada. I expect that with experience, some of the constraints will be eliminated from the regulations.

What are the characteristics of wild plants that may prove of value in breeding projects? Do you have plans to collect feral/wild seed and to breed with them to make new varieties?

Wild plant germplasm is essential to the improvement of crops, and hemp is no exception. Feral hemp in North America, popularly known as "ditchweed" in the US, is derived mostly from low-THC fiber cultivars grown before authorized cultivation ceased in 1945. These plants are of particular value to North Americans because natural selection has adapted them to local conditions of climate, photoperiod, soil, disease, and herbivory. Despite widespread efforts at eradication, populations still persist in North America, and deserve to be maintained for crop development. Indeed, we have initiated a program to collect and maintain wild North American hemp seed, and the possibility is being considered to use these for breeding purposes. Old World wild hemp populations have existed for millennia, exhibit a wider range of physiological and morphological features than North American ruderal plants, and so also are of considerable significance as germplasm. For the most part, structural features of wild plants are much less desirable than the genes for physiological resistance that wild plants contain. Small seeds with thick achenes, as are characteristic of wild hemp seeds, are not desirable for oilseed cultivars, although it is probable that wild plants do have genes that can improve the already impressive nutritive profile of the oil. Irregular branching, short internodes, and considerable wood in the stems, as are characteristic of wild plants, are not desirable for fibre cultivars. Possibly, wild plants could be used directly for some purposes. The exceptional branching and floral production of some wild plants can contribute to the production of essential oils, although this is currently of minor commercial significance. The relative ability of wild plants to withstand no-till or low-till conditions, and the appreciable biomass productivity of some populations, suggests biomass production possibilities, although these may be limited. In large areas of North America, particularly wetlands supporting wild bird populations, so-called "ecovars" (selections from wild populations) of various species are being used for revegetation purposes, and hemp may well be ideal in some situations.
 

torontomeds

Well-Known Member
Pt 4
Do you know of any archeological record of pre-Columbian hemp in Canada or elsewhere in the New World?

This is a sensitive topic, since some Native Americans have contended that they have utilized Cannabis sativa as a source of "hemp fibre" since pre-Columbian days, and therefore should be exempted from current laws. Certainly almost all authorities subscribe to the belief that hemp is a native of Eurasia, not elsewhere. Curiously, the same was said about the hop plant (Humulus lupulus, the source of flavoring for beer), the nearest relative of Cannabis sativa. Until a few decades ago, it was believed that all wild hop plants growing in North America were simply escapes from cultivated hops imported to North America from Europe. However, fossil pollen has definitively shown that the hop plant was present in North America before mankind, and moreover my own research has shown that there are three distinctive varieties of the hop plant that occur only in North America. No one has shown the presence of pre-Columbian pollen of Cannabis sativa in North America, suggesting strongly that this species was introduced to the New World. The belief that Cannabis sativa was present in pre-Columbian North America likely originates from confusion over the ambiguous name "Indian hemp". The name Indian hemp has been loosely applied to the drug class of Cannabis sativa (i.e. subsp. indica), an allusion to the historical narcotic use in India. However, the name Indian hemp is also employed for the North American Apocynum cannabinum, which was used by native North Americans as a textile fibre plant. While all this is suggestive that Apocynum, not Cannabis, is the correct identification of ancient North American woven materials identified as "hemp" the situation has not yet been entirely clarified. There is currently considerable debate over the possibility of pre-Columbian voyages having resulted in intercontinental exchange of crops, and it is conceivable that Cannabis sativa (even if it is indigenous only to Eurasia) was transported to the New World well before Columbus, and indeed was adopted by native North American Indians. To settle the question, the fibre in authentic pre-Columbian North American "hemp" garments should be identified.

So far, Canadian hemp crops have suffered little from pest damage. Do you expect industrial hemp crops to have pest problems in the future?

Hemp has a reputation for resistance to biological attack, although it has been documented that, like virtually all other plant species, it is subject to a wide diversity of insects and diseases. Nevertheless, most disease and pest problems tend not to be severe. Problems are likely to become more apparent in Canada as acreage increases, so that there will be a need to monitor closely and perhaps initiate breeding to combat harmful species that become especially significant. Organically grown hempseed is particularly desirable for market reasons in North America, and therefore it is especially important to develop the industry to minimize use of biocides.

Where will hemp be in twenty years as a industrial crop? Will Canada have an extensive hemp infrastructure in place?

Most new crops achieve only limited or no success. However, hemp may be different. It really is an old crop, that has been excluded from North America only partly because of the natural operation of market conditions; the bad reputation of drug forms of the species obviously has been a roadblock to success. New and exciting applications have been demonstrated in Europe, and at least the European level of success can be achieved in North America, although it is not likely that the European level of crop subsidies will be provided. There is a genuine excitement and enthusiasm on the part of Canadian and indeed American entrepreneurs that I have not seen matched for any other agricultural plant. Research is a key to future success, and it will be essential to increase productivity, both by development of superior cultivars, and by processing technology. I do believe that Canada will have a well developed hemp industry in the next decades. There are, however, two key issues that will determine the extent of this industry. First is the issue of how much THC is permitted in grain (i.e. seed) preparations intended for human food use, a critical issue since salad oils with healthy profiles have huge profit potential. Currently in Canada, food derivatives of the seeds such as oil and seed cake may contain no more than 10 micrograms/gram of THC (i.e., 0.001%), which is attainable. However, governmental authorities in North America have expressed concern that even this very low level poses health risks, and therefore a much lower level is required. This is not practical, at least with current technology and cultivars. The issue is currently being explored, and represents a considerable threat to the industry in North America. Second, there is the issue of the future of hemp in the United States. The US is Canada?s largest market for hemp products. Should hemp cultivation be authorized in the US, as has occurred in Canada, then competition will be severe from American farmers and processors. On the other hand, should Americans be permitted to cultivate hemp, the considerable tension that now exists over Canadian exports of hemp to the US would be eliminated. I would predict that the longer growing season of the US would favor cultivation for fiber purposes, but that Canada would be very competitive in oilseed production, as is the case for rapeseed (Canola) and soybean.
 

torontomeds

Well-Known Member
it seems Ernest Small might be a bit controversial. He is something I happened to come across.

In the 1970s, the taxonomic classification of Cannabis took on added significance in North America. Laws prohibiting Cannabis in the United States and Canada specifically named products of C. sativa as prohibited materials. Enterprising attorneys for the defense in a few drug busts argued that the seized Cannabis material may not have been C. sativa, and was therefore not prohibited by law. Attorneys on both sides recruited botanists to provide expert testimony. Among those testifying for the prosecution was Dr. Ernest Small, while Dr. Richard E. Schultes and others testified for the defense. The botanists engaged in heated debate (outside of court), and both camps impugned the other's integrity.The defense attorneys were not often successful in winning their case, because the intent of the law was clear.

In 1976, Canadian botanist Ernest Small and American taxonomist Arthur Cronquist published a taxonomic revision that recognizes a single species of Cannabis with two subspecies: C. sativa L. subsp. sativa, and C. sativa L. subsp. indica (Lam.) Small & Cronq.The authors hypothesized that the two subspecies diverged primarily as a result of human selection; C. sativa subsp. sativa was presumably selected for traits that enhance fiber or seed production, whereas C. sativa subsp. indica was primarily selected for drug production. Within these two subspecies, Small and Cronquist described C. sativa L. subsp. sativa var. spontanea Vav. as a wild or escaped variety of low-intoxicant Cannabis, and C. sativa subsp. indica var. kafiristanica (Vav.) Small & Cronq. as a wild or escaped variety of the high-intoxicant type. This classification was based on several factors including interfertility, chromosome uniformity, chemotype, and numerical analysis of phenotypic characters.

Professors William Emboden, Loran Anderson, and Harvard botanist Richard E. Schultes and coworkers also conducted taxonomic studies of Cannabis in the 1970s, and concluded that stable morphological differences exist that support recognition of at least three species, C. sativa, C. indica, and C. ruderalis.For Schultes, this was a reversal of his previous interpretation that Cannabis is monotypic, with only a single species. According to Schultes' and Anderson's descriptions, C. sativa is tall and laxly branched with relatively narrow leaflets, C. indica is shorter, conical in shape, and has relatively wide leaflets, and C. ruderalis is short, branchless, and grows wild in central Asia. This taxonomic interpretation was embraced by Cannabis aficionados who commonly distinguish narrow-leafed "sativa" drug strains from wide-leafed "indica" drug strains.
I would like a chance to speak with him, one of the questions I would ask is this.

"The authors hypothesized that the two subspecies diverged primarily as a result of human selection; C. sativa subsp. sativa was presumably selected for traits that enhance fiber or seed production, whereas C. sativa subsp. indica was primarily selected for drug production. Within these two subspecies, Small and Cronquist described C. sativa L. subsp. sativa var. spontanea Vav. as a wild or escaped variety of low-intoxicant Cannabis, and C. sativa subsp. indica var. kafiristanica (Vav.) Small & Cronq. as a wild or escaped variety of the high-intoxicant type. This classification was based on several factors including interfertility, chromosome uniformity, chemotype, and numerical analysis of phenotypic characters."

But did they smoke it? knowing what we know now, it is just as much about terpenes as THC maybe even more impotent. If they did not smoke the so called "C. sativa L. subsp. sativa var. spontanea Vav. as a wild or escaped variety of low-intoxicant Cannabis" How can they know for sure what kind of effect it may or may not have had, maybe the part about people breeding for drug traits is true, but maybe some of the drug traits were not measurable in a lab back in the 70's and 80's, like if they had only been looking at THC and not lets say β-Myrcene and if they only looked at it in a lab and did not sample it personally how can they know for sure about what they are saying?

Anyone who knows the plant knows that there is a sativa and an indica, sativas grow like hemp they way he described, but anyone who knows them knows they are better smoke then indica, stronger, sometimes "Trippy" and we all know that real hemp is nothing like a sativa in its euphoric effects, so maybe he was wrong in that sense. He is calming that any big tall plants are hemp and do not get you high, but I think he is basing that off lab info not field info.
 

Magenta Thumb

Well-Known Member
"even more impotent"...there's a typo worthy of The Hippy.

Wait a few more years for the DNA data (e.g. Jon Page's work) to sort things out.
 

torontomeds

Well-Known Member
"even more impotent"...there's a typo worthy of The Hippy.

Wait a few more years for the DNA data (e.g. Jon Page's work) to sort things out.
1st off why you tripping on one spelling mistake, i typed up that in like 2 seconds at like 5 am.
2 even with DNA that does not tell everything, you still have to take expressions into account. All you can do with DNA is say the potential of one Clone, but if you have 1000 different phenos with in a line how do you test for variance?
 

OGEvilgenius

Well-Known Member
I don't think there is another substance whose safety (long/short term) has been studied more extensively and often by researchers actively looking for a problem who often fail to find any anyway misleading stories and writing be damned.

Did the Liberals say this?
 

OGEvilgenius

Well-Known Member
i think that's why they say there's no studies. they went in looking for the harms and when they found out there was value, they scrapped the study
They're just lying outright actually. I really think this should be something criminal if you are in power.
 
Top