From the point of view of a dynamic morphology, form is not only the result of process(es) -it /s process. This process may be analyzed in terms of two pairs of fundamental processes: growth and decay, differentiation and dedifferentiation. Each of these processes can be analyzed in terms of various modalities (parameters) and submodalities. This paper deals with those of growth (see Table 1). For the purpose of systematics and phylogenetic reconstruction the modalities and submodalities can be considered dynamic characters that have "states'. Each "state" of such a dynamic character is a more detailed process, hence not static. For example, determinate growth represents a %tate ~ of the dynamic character (or modality) of growth duration.
The processes of Table 1 can be applied to the whole plant kingdom (although in certain cases only some processes of the whole set may be applicable). Thus, the diversity of plant form is seen as a diversity of process combinations. From this point of view, change in form implies change in the process combination(s). Questions that arise are, for example, the following: Which process combinations actually occur? Which of these are the most frequent? How and why have process combinations changed during ontogeny and phylogeny?
In comparative morphogenesis, process combinations are compared within an ontogeny or between ontogenies. The combinations may be repeated (i.e., conserved) or changed. Since repetition is limited, regularity that is the basis for structural categories is also limited or relative. With regard to change in process combinations, sequential change within an ontogeny and phylogenetic change between ontogenies can be distinguished. A large number of additional processes, such as beterochrony, that have been investigated by many zoologists and botanists, refer to these sequential and phylogenetic changes.
General implications and consequences of the proposed approach are pointed out. As well, its limits, which are related to the language and concepts used, are discussed. The importance of a dynamic language is emphasized.
RI~SUMI~
La forme des plantes non seulement est le r6sultat de processus, mais aussi le(s) processus m~me(s). En analysant on peut distinguer deux paires de processns fondamentaux: la croissance et la d6composition, la differenciation et la d6diff6renciation. En ce qui concerne la croissance, on peut distinguer en plus un nombre de modalit6s et submodalit6s, chacune avec des "6tats" qui representent des processus plus d6taill6s (Table 1). Par exemple, la submodalit~ de la sym6trie a des 6tats de sym6trie radiale et dorsiventrale qui representent des processus d6taill6s de croissance radiale et dorsiventrale.
De ce point de vue, la diversit6 des formes du r~gne v6g6tale est une diversit6 de combinaisons de ces processus. Certaines des combinaisons sont fr6quentes tandis que d'autres sont rares.
En morphogen~se compar6e on peut distinguer des processus additionels (comme, par example la n6ot6nie) qui d6signent la transformation des combinaisons de processus. Cette transformation peut arriver pendant l'ontogen6se et la phylogen~se.
Plusieurs implications et cons6quences ainsi que des limites de l'approcbe propos6e sont discut6es. I.r.,s limites sont attribuables aux notions et au language utilis6s. L'importance d'un language bas6 sur des notions dynamiques est soulign6e.