Land, the stuff they're not making any more, offers the biggest potential gains and the highest constant risks of any real estate investment medium open to Washingtonians.

Well-located, sensibly-purchased raw land or farmland in the outlying areas of northern Maryland, West Virginia or Virginia - within 100 to 175 miles of downtown Washington - often can double or triple in value within a few years.

The value of some highly speculative tracts, bought in expectation of favorable zoning changes, subdivision development or other commercial use, can jump from a few hundred dollars an acre to $5,000, $10,000 or $20,000 an acre in a month.

Realty brokers and consultants specializing in land in the Washington region can bend your ears with incredible but true stories of fortunes made from forest and farm acreage.

Paul Clark, head of the land commercial realty office of Wolfe, Matan and Sheehan in Frederick, Md., tells about the small group of Washingtonians who bought a 200-acre farm near Urbana, Md., a couple of years ago for $59,000 and leased it out to local farmers for short-term income. The group just sold the farm for $200,000.

Clark's firm, in turn, is subdividing the tract into four plots selling for a total of $260,000 - just a few weeks after the investor group disposed of it.

When you consider that the investors didn't put down anything near the $59,000 sales price - they borrowed money to finance the balance - their return on capital was truly astronomical.

"That's what you're after ideally in land" - the big gain that simply isn't feasible in buying rental houses or second mortgage notes, says Andrew Barr of Barr Real Estate of Arlington.

Barr has been buying chunks of Virginia countryside for years. Among his holdings are a 540-acre tract he describes as an "entire mountain" northwest of Roanoke and an 85-acre parcel of land in Louisa, Va.

The latter cost him $5,500 in November 1960, with $100 down and the remainder financed. He says it has an appraised market value of $100,000 today. What cost him $64.70 per acre 17 years ago would sell fro $1,300 per acre this weekedn, he maintains.

Barr advocates buying low-cost undeveloped land - "not farmland and absolutely not subdivided land, which can be a rip-off" - anywhere within a 150 mile radius of D.C.

"If you hire a local attorney to keep watch on the local real estate agent, and you study what sort of development might be coming 2, 5, 10 years down the road and don't over-commit yourself financially," he contends, "the average investor almost can't lose."

A Washington expert with a less fulsome view of the raw land business is Maury Seldin, head of American University's graduate program in real estate and author of a book on land investment Seldin, who serves as consultant to developers and builders, says his professional experience with land as an investment medium has made him into a conservative on the subject.

"land is at the top of the ladder in terms of risks," he says, "whether you're talking about raw land or farmland. It should be considered very high risk for the amateur investor, who may be unfamiliar with the area and not sure what he's really buying."

Even large-scale developers such as the Levitt building corporation got stung badly on land investments in the early 1970s when the bottom began falling out of the housing market.

A lot of Washingtonians think that as soon as they accumulate a little investment capital they ought to put it into farmland or some mountain acre-age, Seldin notes. But in pure investment terms - return on capital invested vs. risk - rural land may be less advantageous than it looks.

You may not "lose" on a parcel that appreciates at 3 to 5 percent a year for 10 years. But in constant dollars you almost certainly would have done better in a bank account.

Then again, you can buy acreage on which you lose your shirt. You may buy it ridiculously overpriced - such as recreational lots worth a few hundred dollars in real market value that are sold for $3,000 to $5,000 a quarter-acre by unscrupulous promoters.

Land has other inherent challenges. For one thing, it is more difficult to finance than improved real estate. Often an investor's only source of money is his own pocket - which constricts his leverage possibilities - or the seller of the land, which can put an uninformed purchaser in a vulnerable spot. Banks and other institutional lenders typically don't want to lend much on vacant land because it produces no income as security against the debt.

Moreover land offers no immdeiate tax loss benefits because land can't be depreciated, unlike buildings. Unless a buyer finds some interime use for this acreage - such as leasing farm rights to a neighbor, harvesting the timber, renting the land out for campsites, or turning it into a mobile home park - it produces no cash income to offset property tax payments and mortgage service costs.

On top of all this, the notion that raw land values have to go up is simply not ture. "Values go up according to expectations of uses, or actual changes in uses," Seldin points out. A piece of mountain in West Virginia may have no realistic prospect for changed usage in the foreseeable future. Its value per acre therefore has no intrinsic reason for appreciating - unless the Walt Disney people or the Marriott Corp. suddenly want to make a recreational park out of it.

"The point is this," avises Seldin. "Land as an investment can be way oversold," and weekend investors should see the danger signs.

None of which is to say, Seldin hastens to add, that people shouldn't invest in farmland, vacant lots in the suburbs, or central-city lots.

"Quite the reverse," he said. "Land can be the most splendid investment around. It can be the fastest way to get rich - provided you've got your eyes wide open."

Next week: High rolling in the Wastington countryside.

Kenneth R. Harney is editor of the Housing and Development Reporter, published weekly here by BNA, Inc.