This thesis would not have been begun without the two professors who created the Computer Graphics Laboratory at the University of Waterloo and who served as my co-supervisors. When one confronts the decision whether to pursue a PhD, the opportunities to work with and the encouragement from Dr. Kellogg S. Booth and Dr. John C. Beatty (not to be confused with each other!) can not easily be dismissed. The Computer Graphics Laboratory has been a welcome oasis for Canadian computer graphics research and I have enjoyed working there.

The research reported in this thesis was performed at the Xerox Palo Alto Research Center. Xerox has provided exceptional support for this work and has willingly endured my split loyalties between PARC research and PhD research. Maureen Stone and Bill Paxton were responsible for my visiting PARC and later joining the research staff in the Imaging Sciences Laboratory. Several organizational changes caused me to work for Bob Ritchie in the Computer Science Laboratory, where Bob both sheltered me from undue distractions and periodically scrutinized the distractions I was enjoying. PARC has an environment with an incredibly cooperative and collaborative group of computer scientists and the distractions were very tempting. I appreciate my colleagues who patiently waited for my thesis to be finished so that they could work with these tools also.

Maureen Stone and I collaborated on a SIGGRAPH'83 paper that reported the results of the graphical style research presented in Chapter 3. We both share a common interest in high-quality illustrations and her work with Griffin styles evolved independently while I was still in Waterloo.

Bill Paxton implemented Tioga. Tioga is a real jewel in the document composition field and I cannot spend long enough expressing the wonder I felt when I first encountered the Tioga composition system at PARC. It was true pleasure to extend his fine implementation and stretch the boundaries of integrated document composition. Michael Plass designed and implemented the Tioga typesetter and collaborated on the artwork class properties that proved so advantageous in the prototypes I developed. Doug Wyatt crafted the graphics software that serves as the foundation for many of my ideas.

My involvement with the graphic arts and typesetting began in 1974 when Dr. Morven Gentleman convinced the Mathematics Faculty at Waterloo to purchase a small (and cheap) Photon Econosetter. I just could not stay away from experimenting with it when it arrived. Dr. Laurie Rogers, Mark Brader, and Johann George worked on the initial support software; Johann George, Bill Ince, and Alex White collaborated on the initial implementation of TYPESET and its macro packages.

A significant broadening of my appreciation of the graphic arts occurred during 1977 when I collaborated with George Roth, a graphic designer, and John North, a conference chairman, to produce the proceedings for the Third International Conference on Computing in the Humanities (ICCH-3). Over a very short 7-week period, we took 31 conference paper manuscripts, 18 Waterloo co-op students, the Photon Econosetter, and some holiday weekends to produce a hard-bound book in advance of the conference. The manuscripts contained mathematics, computer programs, French, classical Greek, poetry, illustrations, and tables. George Roth very cleverly designed the ICCH-3 proceedings with the considerable discipline required by the limited capabilities of our typesetter and composition software. That initial collaboration grew into a partnership and a company, Waterloo Computer Typography, which subsequently produced several scholarly books and journal articles.

The opportunity of typesetting other people's books became a chance to coauthor an introductory computer science text book. Arnie Dyck, Doug Lawson, and Jim Smith had written a text which I had typeset. When the suggestion arose of my collaborating with them on a revision from WATFIV-S to Pascal, I jumped at the chance. The authors' meetings were very interesting because we had almost total control of the book: graphic design, typography, digital fonts, mathematical typesetting, illustrations, indexing, paste-up. And we were being paid for this learning experience! I am indebted to Arnie Dyck for his attention to detail, insistence on quality, and the great fun we enjoyed teaching together from our book. I marvelled at Doug Lawson for his writing skills and his tutelage in numerical analysis, which surely helped me pass the PhD comprehensive exam.

Finally, I must acknowledge Beth Beach, who is a typographer in her own right, and who made Waterloo Computer Typography the reliable and quality organization that it is. While we typeset several books together, the deficiencies we suffered with our tools motivated the research presented here. I hope this thesis will help make her job easier, both because the research produced some interesting results and because the thesis is finally finished. Now we can enjoy life together without the late hours and concentration devoted to writing the thesis.