This afternoon, as the last frivolous notes of Rossini's "L'Italiana in Algeri" fade into silence in the Terrace Theater, the impresarios of America's major opera companies will settle down for serious business at the Ramada Renaissance Hotel. The general directors of some 70 organizations, ranging from the San Diego Opera to Toronto's Canadian Opera, will gather at the 15th annual conference of Opera America -- a conference whose theme is "The Future of Opera."

For the people who buy tickets, opera is glamorous and exotic; for the members of Opera America, it is a business -- one that is automatically expected to lose money at the box office.

Opera America is a service organization for opera companies, and it concentrates largely on dollars-and-cents topics: how to sell more tickets; how tax changes will affect contributions; who is looking for an artistic director, a marketing manager or a development director. Its newsletter is full of little notices about job openings, scenery and costumes for sale or rent, even sets of surtitles, which have hit American opera like a benign tornado.

But opera blends art with business and so does Opera America. Two days of auditions for singers will precede the business sessions, and while they are in Washington most of the officials will attend current productions of the Washington and Baltimore opera companies.

If we can judge the field by the Washington Opera, host to this year's meeting, opera in America is doing well. The Washington Opera concludes today one of the best seasons -- artistically and economically -- in its 28-year history.

The company has only one serious problem, which it shares with other opera companies: It loses money on every ticket it sells, and it is selling more tickets each year. This season's box office receipts should come close to $2 million, the highest figure in the company's history. But the 1985-86 budget was $5.6 million.

"This means that we have a lot of money to raise," says the company's general director, Martin Feinstein. The budget is considerably less than 10 percent of the Metropolitan Opera's. The Met gives about three times as many performances per year (210) at home, plus another 42 on tour. Its productions feature bigger names than the Washington Opera's, and sometimes they are better. But not that much better.

Ticket sales for the Washington Opera, while higher than ever, are down slightly as a percentage of available seats: approximately 90 percent as compared with about 92 percent last season. This smaller percentage is based on 72 performances, the most the company has ever given in a single season. Feinstein believes that ticket sales for opera in Washington (as for other performing arts) were held down by the fact that it was a presidential election year, when many Washingtonians were too busy to enjoy themselves. The figures look spectacular when they are compared to those for 1980, when the company gave only 16 performances with total ticket sales of $800,000.

Opera was invented in Florence four centuries ago, more or less by mistake, and it has continued in that tradition. A group of poets and musicians that called itself the Cameratawas tried to revive Greek tragedy in its original form. What resulted had some similarities to the original; it was written in verse, it had a chorus, and for the first century or two it was preoccupied with the gods and heroes of antiquity. But it was essentially a new art form and it continued to develop its own quite different identity.

Besides reviving Greek tragedy, the inventors (and occasional reformers) of opera had two other goals. They wanted to enhance the impact of the words in theatrical productions by putting music at their service. And they wanted to achieve a synthesis of all the arts: poetry, music, visual arts and dance. On these goals, opera as a whole has scored reasonably well. The music not only vitalizes the words; it frequently keeps them alive. Many dramas from the past are flourishing as operas that would be hooted off the stage if they were given in spoken form.

In a synthesis, two or more things become one. The perennial quarrel in the operatic synthesis has concerned which one they would become. In France, particularly during the Baroque era, dance and stage spectacle took a dominant role. Even in the 19th century, Giuseppe Verdi and Richard Wagner had to write lavish ballet sequences into their operas if they wanted to succeed in Paris.

The struggle between words and music has been chronic. Antonio Salieri (of "Amadeus" fame) summed up one extreme position, dominant in the late Baroque era, with the title of one of his operas: "Prima la musica; dopo le parole" ("Music first, then words.") This was vigorously opposed by Christoph Gluck at that time, and by Wagner (who wrote his own words) a century later. Wagner's reform was directed against the style known as "bel canto" ("beautiful song"), identified particularly with the music of Gioacchino Rossini, Vincenzo Bellini and Gaetano Donizetti.

For English-speakers, the words-versus-music controversy has a serious complication: the words are usually in a foreign language. If the music doesn't sell tickets, then what will? But a change has been introduced since the arrival of opera on television with subtitles; people have become used to understanding what the pretty sounds all mean. To attract and hold this audience, opera companies all over America have begun using surtitles, flashed on a screen above the stage -- art imitating technology. The Washington Opera began to use titles in some productions this year, with an overwhelmingly positive audience response. Words may be making a comeback.

Also making a comeback is opera as a visual experience -- spurred, like the use of titles, by television. In this, the Washington Opera has been among the prime movers; every production in recent seasons has put a strong emphasis on visual and theatrical values. The choice of the stage director has become as important as that of the prima donna, and the company has scored well in this department. It has no problem choosing a designer; Zack Brown has been doing sets and costumes for almost all its productions recently, and he has done extremely well. His "Rigoletto," "Merry Widow" and "La Bohe me" compare favorably to those of any opera company in the world -- and the list could be made much longer.

Besides Brown, the most notable designer for the company in the last two years has been Jean-Pierre Ponnelle, who designs its joint international productions. The Washington Opera became an international company in the 1983-84 season with a "Cosi fan tutte" that was coproduced with L'Orchestre de Paris and conducted by Daniel Barenboim. It was the first installment of a Mozart trilogy produced jointly in Washington and Paris -- followed this season by "The Marriage of Figaro" and next season by "Don Giovanni." Meanwhile, the Washington Opera for the first time exported one of its own productions to Europe: Gian Carlo Menotti's "The Medium" and "The Telephone" (designed, superbly, by Brown), which opened last summer's Edinburgh Festival and will be featured later this year at a festival in Israel.

In spite of this international recognition, that production (one of the best the company has ever done) accounted for a substantial part of the empty seats this season, partly because it was running for its second consecutive year, but also because there is a limited audience for modern operas -- even by Menotti. Those who missed the performance of Beverly Evans in "The Medium" missed one of the most intense theatrical experiences of our time.

On the words-versus-music front, it should be noted that more and more operas are being written with words in English. Of the five Washington and Baltimore operas being shown while Opera America is meeting here, only one, "L'Italiana in Algeri," is in a foreign language. American composers today have the talent to generate a golden age of opera comparable to any in history; what they lack is an audience hungry for new operas.

Whether or not it has mass appeal, modern opera is one of the things the company does best -- particularly in the small, acoustically superb Terrace Theater. This was demonstrated in Igor Stravinsky's "The Rake's Progress," a revival that began well, survived two crucial changes of cast and ended the season stronger than ever. But another revival, "La Bohe me," was probably the most popular operatic event of the season.

The use of three revived productions is, in a sense, a sign of the company's coming of age. A few years ago it owned no productions of its own; now it is able to bring back productions, earning added returns on the capital investments it made for sets and costumes. The norm in the past few years has been to have two revived productions -- one carried over from the previous year and one from several years earlier. Besides eliminating a large start-up expense for revivals, the stock of sets and costumes can be rented to other companies, a small but not insignificant source of income.

The future of opera, as the opera impresarios of America sit down to consider it today, looks complicated (opera is always complicated) but positive. Deficits may be growing, but so are audiences. Opera is an old form and, on the whole, a conservative one, but it is adapting to changing conditions with such innovations as surtitles. Opera companies tend to be solitary phenomena; most cities each have only one worth mentioning. But they are aware that they have many things in common beyond the occasional joint production or rental of scenery. That's why they are getting together to talk about their future, and that they are discussing it should help to improve it.