In this multidisciplinary glimpse forward, some of this decade's key players offer opinions on a range of topics- from what has driven progress, to where innovation will come from and to obstacles we have yet to overcome. In this excerpt from Visions for the Future of the Fields, a panel discussion held at the 10th anniversary of the US Computer Science and Telecommunications Board, experts identify critical issues for various aspects of computing. In the accompanying sidebars, some of the same experts elaborate on points in the panel discussion in mini-essays: David Clark, CSTB chairman, looks at the changes needed, as computing science research comes of age. The current context of computer science, Clark states, is shaped by past successes and chronic trouble spots. Mary Shaw, Carnegie Mellon University, examines challenges for software system designers. Shaw states that disintermediation-the direct connection of users and their software-has created new problems for software system designers. Robert Lucky, Bellcore, looks at IP dial tone, a new infrastructure for the Internet. Lucky states that data traffic will soon eclipse voice traffic, portending not just a revolution in technology but in the very basis of telecommunications economics. Donald Greenberg, Cornell University, rounds out the essays with an outlook on computer graphics. Greenberg states that by 2025 we will have the technology to produce realistic real-time images at resolutions up to or beyond the limits of our visual perception.Finally, in an interview with William Wulf, president of the US National Academy of Engineering, Computer explores the roots of innovation and the broader societal aspects that will ultimately drive innovation in the near term.

The Ventilator Manager (VM) program is an experiment in expert system development that builds on our experience with rules in the MYCIN system. VM is designed to interpret on-line quantitative data in the intensive care unit (ICU) of a hospital. After a major cardiovascular operation, patient often needs mechanical assistance with breathing and is put in the ICU so that many parameters can be monitored. Many of those data are relevant to helping physicians decide whether the patient is having difficulty with the breathing apparatus (the ventilator) or is breathing adequately enough to remove the mechanical assistance. The VM program interprets these data to aid in managing postoperative patients receiving mechanical ventilatory assistance. VM was strongly influenced by the MYCIN architecture, but the program was redesigned to allow for the description of events that change over time. VM is an extension of a physiologic monitoring system1 and is designed to perform five specialized tasks in the ICU:

Artificial intelligence research, as a part of computer science, has produced a variety of programs of experimental and applications interest: programs for scientific inference, chemical synthesis, planning robot control, extraction of meaning from English sentences, speech understanding, interpretation of visual images, and so on. The symbolic manipulation techniques used in artificial intelligence provide a framework for analyzing and coding the knowledge base of a problem independently of an algorithmic implementation. A possible application of artificial intelligence methodology to protein crystallography is described.

ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTApplications of artificial intelligence for chemical inference. 22. Automatic rule formation in mass spectrometry by means of the meta-DENDRAL programB. G. Buchanan, D. H. Smith, W. C. White, R. J. Gritter, E. A. Feigenbaum, J. Lederberg, and Carl DjerassiCite this: J. Am. Chem. Soc. 1976, 98, 20, 6168–6178Publication Date (Print):September 1, 1976Publication History Published online1 May 2002Published inissue 1 September 1976https://pubs.acs.org/doi/10.1021/ja00436a017https://doi.org/10.1021/ja00436a017research-articleACS PublicationsRequest reuse permissionsArticle Views297Altmetric-Citations67LEARN ABOUT THESE METRICSArticle Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated. Share Add toView InAdd Full Text with ReferenceAdd Description ExportRISCitationCitation and abstractCitation and referencesMore Options Share onFacebookTwitterWechatLinked InRedditEmail Other access optionsGet e-Alertsclose Get e-Alerts

Chemischer InformationsdienstVolume 7, Issue 52 Physical Organic Chemistry ChemInform Abstract: APPLICATIONS OF ARTIFICIAL INTELLIGENCE FOR CHEMICAL INFERENCE. 22. AUTOMATIC C RULE FORMATION IN MASS SPECTROMETRY BY MEANS OF THE META-DENDRAL PROGRAM B. G. BUCHANAN, B. G. BUCHANANSearch for more papers by this authorD. H. SMITH, D. H. SMITHSearch for more papers by this authorW. C. WHITE, W. C. WHITESearch for more papers by this authorR. J. GRITTER, R. J. GRITTERSearch for more papers by this authorE. A. FEIGENBAUM, E. A. FEIGENBAUMSearch for more papers by this authorJ. LEDERBERG, J. LEDERBERGSearch for more papers by this authorC. DJERASSI, C. DJERASSISearch for more papers by this author B. G. BUCHANAN, B. G. BUCHANANSearch for more papers by this authorD. H. SMITH, D. H. SMITHSearch for more papers by this authorW. C. WHITE, W. C. WHITESearch for more papers by this authorR. J. GRITTER, R. J. GRITTERSearch for more papers by this authorE. A. FEIGENBAUM, E. A. FEIGENBAUMSearch for more papers by this authorJ. LEDERBERG, J. LEDERBERGSearch for more papers by this authorC. DJERASSI, C. DJERASSISearch for more papers by this author First published: December 28, 1976 https://doi.org/10.1002/chin.197652060AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat No abstract is available for this article. Volume7, Issue52December 28, 1976 RelatedInformation

Chemischer InformationsdienstVolume 5, Issue 2 Physical Organic Chemistry ChemInform Abstract: APPLICATIONS OF ARTIFICIAL INTELLIGENCE FOR CHEMICAL INFERENCE PART 10, INTSUM, A DATA INTERPRETATION AND SUMMARY PROGRAM APPLIED TO THE COLLECTED MASS SPECTRA OF ESTROGENIC STEROIDS D. H. SMITH, D. H. SMITHSearch for more papers by this authorB. G. BUCHANAN, B. G. BUCHANANSearch for more papers by this authorW. C. WHITE, W. C. WHITESearch for more papers by this authorE. A. FEIGENBAUM, E. A. FEIGENBAUMSearch for more papers by this authorJ. LEDERBERG, J. LEDERBERGSearch for more papers by this authorCARL DJERASSI, CARL DJERASSISearch for more papers by this author D. H. SMITH, D. H. SMITHSearch for more papers by this authorB. G. BUCHANAN, B. G. BUCHANANSearch for more papers by this authorW. C. WHITE, W. C. WHITESearch for more papers by this authorE. A. FEIGENBAUM, E. A. FEIGENBAUMSearch for more papers by this authorJ. LEDERBERG, J. LEDERBERGSearch for more papers by this authorCARL DJERASSI, CARL DJERASSISearch for more papers by this author First published: January 15, 1974 https://doi.org/10.1002/chin.197402065Read the full textAboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onEmailFacebookTwitterLinkedInRedditWechat No abstract is available for this article. References D. H. SMITH, B. G. BUCHANAN, W. C. WHITE, E. A. FEIGENBAUM, J. LEDERBERG, CARL DJERASSI, APPLICATIONS OF ARTIFICIAL INTELLIGENCE FOR CHEMICAL INFERENCE PART 10, INTSUM, A DATA INTERPRETATION AND SUMMARY PROGRAM APPLIED TO THE COLLECTED MASS SPECTRA OF ESTROGENIC STEROIDS, Tetrahedron, 1973, 29, 3117. DOI: 10.1016/S0040-4020(01)93453-2; 10.1016/S0040-4020(01)93453-2 CASWeb of Science®Google Scholar Volume5, Issue2January 15, 1974 ReferencesRelatedInformation

A method for systematic interpretation and summary of evidence found for all possible mass spectral fragmentations of a molecule or set of related molecules is described. The method is embodied in a computer program (INTSUM) which interprets, in terms of fragmentation processes, mass spectral data collected on known compounds. Utilizing high resolution mass spectra from 47 estrogenic steroids, the method is verified and new findings are discussed. Finally, the method is used to explore the fragmentations of equilenins and several acetate and benzoate ester derivatives.

A method for systematic interpretation and summary of evidence found for all possible mass spectral fragmentations of a molecule or set of related molecules is described. The method is embodied in a computer program (INTSUM) which interprets, in terms of fragmentation processes, mass spectral data collected on known compounds. Utilizing high resolution mass spectra from 47 estrogenic steroids, the method is verified and new findings are discussed. Finally, the method is used to explore the fragmentations of equilenins and several acetate and benzoate ester derivatives.

Chemischer InformationsdienstVolume 3, Issue 45 Isocyclic Compounds ChemInform Abstract: ANWENDUNG KUENSTLICHER INTELLIGENZ FUER CHEMISCHE SCHLUSSFOLGERUNGEN 8. MITT. DEUTUNG HOCHAUFGELOESTER MASSENSPEKTREN KOMPLEXER MOLEKUELE MITTELS RECHENAUTOMATEN, STRUKTURAUFKLAERUNG OESTROGENER STEROIDE D. H. SMITH, D. H. SMITHSearch for more papers by this authorB. G. BUCHANAN, B. G. BUCHANANSearch for more papers by this authorR. S. ENGELMORE, R. S. ENGELMORESearch for more papers by this authorA. M. DUFFIELD, A. M. DUFFIELDSearch for more papers by this authorA. YEO, A. YEOSearch for more papers by this authorE. A. FEIGENBAUM, E. A. FEIGENBAUMSearch for more papers by this authorJ. LEDERBERG, J. LEDERBERGSearch for more papers by this authorCARL DJERASSI, CARL DJERASSISearch for more papers by this author D. H. SMITH, D. H. SMITHSearch for more papers by this authorB. G. BUCHANAN, B. G. BUCHANANSearch for more papers by this authorR. S. ENGELMORE, R. S. ENGELMORESearch for more papers by this authorA. M. DUFFIELD, A. M. DUFFIELDSearch for more papers by this authorA. YEO, A. YEOSearch for more papers by this authorE. A. FEIGENBAUM, E. A. FEIGENBAUMSearch for more papers by this authorJ. LEDERBERG, J. LEDERBERGSearch for more papers by this authorCARL DJERASSI, CARL DJERASSISearch for more papers by this author First published: November 7, 1972 https://doi.org/10.1002/chin.197245085AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat No abstract is available for this article. Volume3, Issue45November 7, 1972 RelatedInformation

Chemischer Informationsdienst. Organische ChemieVolume 2, Issue 6 Organoelement Compounds ChemInform Abstract: ANWENDUNG VON ELEKTRONENRECHNERN AUF CHEMISCHE PROBLEME 4. MITT. GESAETTIGTE AMINE, DIE DURCH MASSENSPEKTREN GERINGER AUFLOESUNG UND NMR-SPEKTREN GEKENNZEICHNET SIND ARMAND BUCHS, ARMAND BUCHSSearch for more papers by this authorA. M. DUFFIELD, A. M. DUFFIELDSearch for more papers by this authorGUSTAV SCHROLL, GUSTAV SCHROLLSearch for more papers by this authorCARL DJERASSI, CARL DJERASSISearch for more papers by this authorALLAN B. DELFINO, ALLAN B. DELFINOSearch for more papers by this authorB. G. BUCHANAN, B. G. BUCHANANSearch for more papers by this authorG. L. SUTHERLAND, G. L. SUTHERLANDSearch for more papers by this authorE. A. FEIGENBAUM, E. A. FEIGENBAUMSearch for more papers by this authorJ. LEDERBERG, J. LEDERBERGSearch for more papers by this author ARMAND BUCHS, ARMAND BUCHSSearch for more papers by this authorA. M. DUFFIELD, A. M. DUFFIELDSearch for more papers by this authorGUSTAV SCHROLL, GUSTAV SCHROLLSearch for more papers by this authorCARL DJERASSI, CARL DJERASSISearch for more papers by this authorALLAN B. DELFINO, ALLAN B. DELFINOSearch for more papers by this authorB. G. BUCHANAN, B. G. BUCHANANSearch for more papers by this authorG. L. SUTHERLAND, G. L. SUTHERLANDSearch for more papers by this authorE. A. FEIGENBAUM, E. A. FEIGENBAUMSearch for more papers by this authorJ. LEDERBERG, J. LEDERBERGSearch for more papers by this author First published: February 9, 1971 https://doi.org/10.1002/chin.197106071Read the full textAboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat No abstract is available for this article. Volume2, Issue6February 9, 1971 RelatedInformation

: A computer program has been written which can formulate hypotheses from a given set of scientific data. The data consist of the mass spectrum and the empirical formula of an organic chemical compound. The hypotheses which are produced describe molecular structures which are plausible explanations of the data. The hypotheses are generated systematically within the program's theory of chemical stability and within limiting constraints which are inferred from the data by heuristic rules. The program excludes hypotheses inconsistent with the data and lists its candidate explanatory hypotheses in order of decreasing plausibility. The computer program is heuristic in that it searches for plausible hypotheses in a small subset of the total hypothesis space according to heuristic rules learned from chemists. (Author)

A method for systematic interpretation and summary of evidence found for all possible mass spectral fragmentations of a molecule or set of related molecules is described. The method is embodied in a computer program (INTSUM) which interprets, in terms of fragmentation processes, mass spectral data collected on known compounds. Utilizing high resolution mass spectra from 47 estrogenic steroids, the method is verifiedand new findings are discussed. Finally, the method is used to explore the fragmentations of equilenins and several acetate and benzoate ester derivatives. Interpretationof the mass spectra of known compounds to elucidate mechanisms of fragmentation has been performed manually for many years. The results of such interpretations have resulted in a significantbody of empiricalrules relatingfreatures of molecular structures to the ways in which the structures fragment subsequent to ionization. When more than a few related structures are involved, the task of manual examination of the spectra to determine related modes of fragmentation can become complex and tedious. The advent of high resolution mass spectra to limit compositional ambiguities inherent in low resolution mass spectral data has, to some extent, eased the problemof interpretation.Several techniqueshave been suggested which are designed to present high resolution spectra in a format that aids interpretation.'"' It can be argued, however, that detailedmechanistic interpretationof data in a single high resolution spectrum, and, particularly, comparison of several spectra of related compounds is still a difficult task utilizing these techniques. This particular area of data interpretation seems well suited for study by techniques of artificial intelligenceusing the heuristic search paradigm." The search in this case is over the space of possible fragmentation processes. The heuristics, or rules, +This paper is dedicated to Professor Edgar Lederer of Institut de Chimie des Substances Naturelles. Gif-suron the occasion of his sixty-fifth birthday. +For Part IX, see D. H. Smith, B. G. Buchanan, R. S. Engelmore, H. Adlercreutz and C. Djerassi, /. Am. Chem. Soc. submitted. §Derivatives of the l,3,5(10)-estratriene skeleton. used to guide the search are those empirical rules about fragmentation probabilities whicjh are used routinely in manual data interpretation.Theserules are sometimes very good, but sometimes inadequate. The use of such judgemental knowledge without guarantees of success characterizesartificial intelligence programs as a whole. Interpretation of large amounts of data is a task well suitedfor a computer. It can explore the space of possible interpretations much more systematically and tirelessly than a chemist can. Computer techniques related to this approach have recently been utilized in mechanistic interpretation of individual low resolution mass spectrum/structure pairs.5 A set of65 completehighresolution mass spectra of estrogenic steroids§ was available to test theperformance of the computer program (termed INTSUM) written for data interpretation and summary. The first goal was to verify the performance of the program by comparing its results with manual interpretation, using a set of 47 compounds closely related to those studied previously." This verification step offered the advantage of ensuring that generalizations developed previously, utilizing low resolution mass spectra, were correct. This step also offered the possibility of investigation,across a widevariety of compounds,of certain fragmentation processes which were of limited generality when incorporated into a program for automatic structure elucidation.' The remainder of the set of compounds was utilized to fulfill the second goal, namely, use of INTSUM to explore fragmentation processes for compounds (equilenins and several acetateandbenzoate esters) 3117 (Received in USA 17 January 1973;Received in UKfor publication 29 May 1973) 3118 D. H. Smith et al. whosemass spectra have been subject to little or no prior investigation. The program is described in the context of operation with high resolution mass spectral data. It is capable of analysis of low resolution mass spectra, with possible elemental compositions for each fragment ion limited only by the nominal mass and the heteroatomcontent of the molecule. As with manual interpretation there is a considerable increase in ambiguityof explanatoryhypotheses when low resolution spectra are analyzed. The philosophy underlying development of this program, its importance to automatic theory formation, a long-range goal of this research, and an overview of requirements for generality and thoroughness have been presented."

An extension of the Heuristic dendral program has been used for automatic interpretationof the complete high resolution mass spectra of estrogenic steroids. The program has been structured for facile adaptationto thegeneralproblem ofanalysis ofthe highresolution mass spectraofother classes ofcomplex organic compounds. The operationof the program and its performancein interpretationof the spectra of 43 estrogenrelated compoundsare described. Its performance is comparable to the quality and surpasses in speed theperformance of trained mass spectroscopists. Intelligent computer programs for chemical inference about structure elucidation from mass spectral and other data have yielded promising results. lb -3 The Heuristic dendral program interprets the low resolution mass spectra of a large variety of saturated, aliphatic, monofunctional compounds in terms of molecular structure. Other approaches to analysis of low resolution mass spectra have included library matching procedures, recently reviewed by Hertz, et al.,* procedures involving a combination ofmatching and interpretation, 5'6 and totally empirical learning machines.7 In this paper we introduce the systematic use of high resolution mass analysis and of metastable ions and their application to a more complex set of structures, the estrogenicsteroids. The interpretive methods used in Heuristic dendral differ from these other methods in several respects: (a) the Heuristic dendral program uses a set of empirical mass spectrometryrules (hereafter referred to as a "theory") which is nearly comparable in complexity to the theory used by human chemists; (b) the program's process of reasoning from data to explanation, using the theory of mass spectrometry, emulates the idealized reasoning process of. an experienced mass spectroscopist; (c) the Heuristic dendral program is capable of examining all possible molecular structures (1) (a) Financial support for these investigations was provided in part by the National Institutes of Health, Grant No. RR-612-01, the Advanced Research Projects Agency, Grant No. SD-183, and the National Aeronautics and Space Administration, Grant No. NGR-005-020-004; (b) for part VI, see A. Buchs, A. B. Delfino, A. M. Duffield, C. Djerassi, B. G. Buchanan, E. A. Feigenbaum, and J. Lederberg, Helv. Chim. Acta, 53, 1394(1970). (2) For part VII, see A. Buchs, A. Delfino, C. Djerassi, A. M. Duffield, B. G. Buchanan, E. A. Feigenbaum, 1. Lederberg, G. Schroll, and G. L. Sutherland, Adi-an. Mass Spectrom., 5, 314 (1971). (3) (a) A. M. Duffield, A. V. Robertson, C. Djerassi, B. G. Buchanan, G. L. Sutherland, E. A. Feigenbaum, and J. Lederberg, J. Amer. Chem. 91, 2977 (1969); (b) A. Buchs, A. M. Duffield, G. Schroll, C. Djerassi, A. B. Delfino, B. G. Buchanan, G. L. Sutherland, E. A. Feigenbaum, and J. Lederberg, ibid., 92, 6831 (1970); (c) G. Schroll, A. M. Duffield, C. Djerassi, B. G. Buchanan, G. L. Sutherland, E. A. Feigenbaum, and J. Lederberg, ibid., 91, 7440 (1969). (4) H. S. Hertz, R. A. Hites, and K. Biemann, Anal. Chem., 43, 681 (1971). (5) L. R. Crawfordand J. D. Morrison, ibid., 43, 1790 (1971). (6) D. H. Smith, ibid., 44, 536 (1972). (7) P. C. Jurs, 43, 1812 (1971), and references cited therein. within a class of compounds, or fitting an observed molecular formula, in order to choose the best explanation of the data; (d) the program is able to reduce the number of structures actually considered by referencing the data, including other data such as nmr, when available. Usingrules ofmass spectrometry developedfrom a small set of spectra of standard compounds, the Heuristic dendral program is able to analyze large numbers of mass spectracorrectly. Because the Heuristic dendral program uses a theory of mass spectrometry in much the same way that mass spectroscopists do, it is possible for chemists to understand the reasoning steps of the program. Thus they can suggest extensions or alternative steps when the program fails to analyze some spectra correctly (a difficulty of some statistical approaches7). This is a great advantage for building a powerful program in incremental steps. Research in most disciplines of organic chemistry involves complex, generally polyfunctional molecules that cannot be analyzed easily by existing computer programs. One direction of future computer research in chemistry will certainly be to develop programs that reason about molecules of greater complexity. Considering automated structure elucidation based, at least in part, on mass spectral data, the multiplicity of possible elemental compositions in low resolution mass spectra of complex molecules is an effective deterrent to the successful implementation of the various approaches mentioned above. It is apparent that the specificity of complete high resolution mass spectra, where elemental compositions of all ions are determined, is required.*^9b High resolution mass spectra have been utilized in conjunction with computer programs to aid in the structure elucidation of a limited number of types of com(8) (a) A. Mandelbaum, P. V. Fennessey, and K. Biemann, Proc. Annu. Conf. Mass Spectrom. Allied Topics, 15th, 111 (1967); (b) R. Venkataraghavan, F. W. McLatt'erty, and G. E. VanLear, Org. Mass Spectrom., 2, 1 (1969); (c) W. J. Richter, B. R. D. H. Smith, and A. L. Burlingame, Anal. Chem., 41, 1392 (1969). (9) (a) M. Senn, R. Venkataraghavan, and F. W. McLafferty, /. Amer. Chem. Soc, 88, 5593 (1966); (b) X.. Biemann, C. Cone, B. R. Webster, and G. P. Arsenault, ibid., 88, 5598 (1966). [Reprinted fromthe Journal ofthe AmericanChemical Society, 94, 5962 (1972) ] Copyright 1972 by theAmerican Chemical Societyand reprinted bypermission ofthecopyright owner. 5963 pounds,8 most notably, peptides.9 A submolecular group analysis method, wherein certain combinations ofelements are assumed to represent certain portions of molecules, has been proposed as a more general approach. 10 However, there has been little effort toward a systematicapproach to computerized interpretation of high resolution mass spectra. The availability of the facilities for structure generationand manipulation in the dendral algorithm, ll and the ability to encode knowledge about mass spectral fragmentation processes (heuristics, rules) in a simple format that can be handled by the computer, permit an implementation of such a systematic approach. As will be shown in a subsequent publication, utilization of metastable ion data can permit the analysis of mixtures of compounds without prior separationof the constituents. The program described below is an extension of the Heuristic dendral program previously mentioned. Given information about the basic structural unit of a class of compounds, the fragmentation mechanisms general to that class, a high resolution mass spectrum and metastable ion information, the program attempts to determine a molecular structure(s) to explain the data. One will note that this program is tied more rigidly to the data than the chemist, since it cannot bring to bear on the problem chemical experience for which it has not been programmed. On the other hand, the program is often much more thorough in its systematic search through the data and in its consideration of all combinations of evidence uncovered by it. The present approach differs from the method of the Heuristic dendral program in at least two important respects. First, this work does not encompass a systematic program for enumerating molecular structures comparable to the dendral acyclic structure generator. 11 The reason for this is that no structure-generating program of sufficient scope and flexibility has yet been written. Second, this work does not encompass a predictor like that of the Heuristic dendral system which allows ranking the final structures Essentially, the work described here is anextension of the Heuristic dendralsystem's planning phase. High Resolution Mass Spectrum Metastable Peaks