1. 3 different approaches can apply to the problems to which the rule of first country of asylum responds: a) From a humanitarian point of view the refugee should be allowed to choose his country of asylum freely, and no refugee should be returned with reference to the rule of first country of asylum. b) If the refugee is returned without security of asylum elsewhere, his "right" to asylum is violated. This creates orbit-refugees and refoulement risks, even if this consequence is not intended by the rejecting rountry. c) The duties imposed on the first country of asylum, may violate the principles of burden sharing. 2. The main reason behind the rule of first country of asylum is the necessity to protect countries of asylum from mass influx. However, it is not likely that certain countries are so much more attractive to great masses of sponteneous refugees. The number of individual and general factors determining each refugee's wishes and the variation of countries make distribution more likely than concentration. Experience in countries with en route provisions confirm this. Furthermore the rule of first country of asylum may lead to mass influx in the "first countries". Another reason, which is not suffient in itself, is the administrative advantages of the rule. The restrictive general aliens policy may have off spring effects upon the refugee policy, leading to widespread use of the rule of first country of asylum. A final reason for the rule may be the wish of states to enforce other states to respect their duties to grant asylum. 3. In particular this last reason illustrates that refugees are used "instrumentally", i.e. to further purposes beyond themselves and their lives. The rejection of refugees without security for protection elsewhere is incompatible with a humanitarian legal order. The application of rules of first country of asylum is not just a rejection of elected asylum. The recognition of elected asylum is a humanitarian, and probobly expedient, conduct. Refusal of elected asylum in cases where protection is secured elsewhere, is certainly less humanitarian. Rules according to which refugees are sent on to an insecure and often dangerous life, are ethically unacceptable and inhuman. Accordingly those who wish to maintain the rule of first country of asylum will have to lift the "burden of argument". 4. Refusal of asylum because asylum should be claimed elsewhere appears in many variations. If rejection takes place only when asylum is aquired elsewhere, the only consequence is the rejection of elected asylum. From this situation deterioation may happen in relation to what has been achieved (e. g. a weaker form of protection) as to the level of proof hereof (e. g. a reasonable assumption that asylum may be achieved). However undesirable, pre-screening has no direct relation to the problems of first country of asylum. The refoulement agreements do not contribute to a solution of this problem. On the contrary they only decide which country does not have the responsibility for an asylum-seeker, but contain nothing about which state should positively take up the asylum case. 5. If all countries had similar procedual and material rules, the only disadvantage of the rule of first country of asylum would be the lack of elected asylum. Whereas it might be possible to harmonize rules of procedure, substantial harmonization of material rules seems unlikely. A harmonization of material rules alone (e. g. with recognition of the rule of first country of asylum) would not in itself solve the problem. On the other hand the problem can be solved without harmonization. 6. Principles of burden-sharing and distribution are today hardly suitable for more than declarations of intent. The only possible solution today seems to be a classification and harmonization of the concept of first country of asylum combined with a suspension of these rules in case of mass influx in a first country of asylum. 7. Comments to Melander's draft convention in "Refugees in orbit": a) The precise definition of de facto refugees implies that certain people or categories would be excluded. This can be avoided by merely referring to a country's practice regarding de facto refugees. b) It is right to exclude de facto refugees from the res judicata provision. c) It is reasonable to apply the "en route" model as the basic criterion for defining the first country of asylum, and a common time-limit is necessary even if individual circumstances may influence the reasonable duration of the "en route" requirements in each case. d) It is suggested to include a provision on burden-sharing comprising: - a right for a first country to denounce its duties with reference to mass influx. - a corresponding duty for another country. - a co-ordinating body. 8. It is suggested to accept the concept of a provisional "common asylum" in Western Europe, combined with the establishment of a European Refugee Commission, to decide on the first country of asylum (but not on the material issues). 9. Unfortunately the most ideal solution: A right for each refugee to choose his country of asylum, remains in the world of ideals for the time being.