The Mathematical Institute of the Serbian Academy of Sciences and Arts was founded in 1946 as the first institute of the Academy. At the time of its founding in a country devastated by the war, the Institute had to start practically from zero, but there was a historical background on which to build.

Scientific research in mathematics began in Serbia near the end of the Nineteenth century. It was performed by a handful of professors of Belgrade University who received their degrees from the best universities in Germany, Austro-Hungary and France. They received a significant boost after the WWI with the arrival of emigrants from Russia, some of whom were prominent mathematicians.

The predecessor of the Institute was the so called "Mathematician's Club", established in the twenties. It gathered all research-oriented mathematicians from Belgrade University, organized monthly lectures (colloquium talks) and, from 1932, started publishing the first international mathematics journal in Serbia:*Publications Mathématique de l'Université de Belgrade*. At their disposal they had a rich collection of materials in the Mathematics Seminar of the Belgrade University, established in 1895. By 1946, however, all this was gone. In 1944, just two days before the liberation, the library was burned together with all the written records of the Club's activities. The last volume of *Publications* was also burned in the printing shop, during the bombardment in 1941, and only one copy remained in the house of the editor.

The founders and the first seven members of the Institute, all former members of the Mathematician's Club, saw the Institute as the best means for rebuilding mathematics in Serbia. They immediately elected another seven members and started with monthly lectures, which soon turned into weekly meetings. The new library was started at the same time and by 1947 *Publications*reappeared, now with a modified name: *Publications de l'Institute Mathématique*.

The most difficult problem, however, was the shortage of young members. The founding members, all very prominent personalities and internationally known scientists, were of quite advanced age, only one being under 60. In 1949 the Institute obtained its first young assistant and, from that time, steady growth continued over the decades.

From the beginning, the Institute had a concept which is very modern today. Rather than employing a large staff, as research institutes frequently do, the idea was to have members employed at other institutions while the Institute provides only the infrastructure for research: the library, colloquiums, seminars and courses, publications etc. Such decision was probably prompted by the specific way in which Universities in Serbia are organized where mathematicians are not grouped in one department but divided into different schools (Faculties). After the rapid growth of the University network in Serbia, the Institute now has collaborators employed at over 30 different institutions, mostly Faculties of the Universities in Belgrade, Novi Sad, Niš, Kragujevac and Pristina.

In the fifties and sixties the Institute always employed a handful of young assistants who were encouraged to transfer to Universities after receiving their degrees. The Institute actually provided important support in creating Mathematics departments at all our Universities. From the seventies, when legislation started mandating the minimum number of permanent members for scientific institutes, the Mathematical Institute had permanently employed scientists, but their number was always kept near the legal minimum. Currently the Institute employs about thirty mathematicians (twenty Ph.D.'s and ten graduate students) out of which about ten are always abroad (as graduate students, visiting professors etc.). At the same time, however, the Institute engages about 250 researchers on its projects with another 150 being more loosely associated with the Institute. As always, these part-time members play a major role in managing the Institute, being members of the Board and Scientific council or taking the positions of project leader, seminar chairman, editor, etc.

The founding fathers had, among them, a colorful combination of degrees in mathematics as well as in astronomy or different kinds of engineering. Consequently, they considered mechanics (including applications in mechanical and civil engineering), astronomy and theoretical physics as parts of applied mathematics.

They and their pupils were in a position to impose on the Technical School of the Belgrade University a very extensive and rigorous mathematical curriculum, which was later copied by all the newly founded Universities in Serbia. Thanks to this, our engineers are now among the mathematically best educated engineers in the world. In fact, quite a few of them are doing research on projects at the Institute, some of them being among the leaders.

In its first two decades, Analysis and its applications in Mechanics absolutely dominated the research activities at the Institute. But in general, the institute was never sectarian: research in any branch of mathematics was always warmly welcomed, as long as it was of high quality. Over the decades various branches had their ups and downs, based mainly on their agility and ability to attract graduate students through seminars and other activities. In the seventies, for example, Logic progressed from virtual nonexistence to the status of the most vigorous group. In the eighties Geometry and Topology became more prominent. Currently, we are witnessing the rejuvenation of Analysis and Mechanics. Scientific policy decisions generally encouraged variety and only occasionally gave special support in cases where groups were falling behind or new disciplines were being established. In that way Mathematics in Serbia not only grew during the past 50 years from less then 20 researchers to more than 400, but also expanded in scope tremendously, covering now practically all branches of modern mathematics and its applications, and thus keeping pace with the explosive development of Mathematics in the second half of the Twentieth Century.

The Institute recognized quite early on that computer science was important for further development of mathematics. In the mid-sixties the Computer Center of the Institute was created and an IBM 360 was acquired, the best scientific computer of the time. After the initial enthusiasm and some interesting results, however, the whole idea slowly deteriorated. One reason was certainly the limited aptitude of mathematicians for managing the Big Science (the upkeep of IBM 360 was comparable to that of a major installation in physics, not to mention eventual replacement). Another reason stemmed from political circumstances. The dissolution of Yugoslav federation began in the late sixties with the dissolution of Federal Scientific Research Fund. The seventies, a time of economic prosperity in Yugoslavia financed by foreign loans, saw the downfall of Yugoslav science.

The industry, as well as the government, abandoned research and development in favor of purchasing or licensing foreign technology, which brought expensive, applied research to a virtual standstill. The advent of personal computers facilitated the decision to close down the Computer Center in 1985 and start building a network of PCs. The interest of the Institute in computer science was redefined at that time to include only those areas of research that can be done on a PC. Fortunately, very soon this came to mean almost everything of some interest for mathematicians.

In 1961 the Institute acquired an independent status but remained, willingly, under the auspices of the Serbian Academy of Sciences and Arts, where it is still housed today.

Zoran Marković, Director