2. "AN INFORMATION STORAGE AND RETRIEVAL SYSTEM 
FOR BIOLOGICAL AND GEOLOGICAL DATA-DESIGN 
CONSIDERATIONS" 
Reginald Creighton 
Smithsonian Institution 
Washington, D.C. 
The system design attempts to achieve optimum 
processing economy while permitting the system user 
maximum freedom of activity regarding data prepara 
tion, query expression and output presentation. 
Due to the disparity of data parameters involved 
(birds, crustecia, and rocks) a completely genera¬ 
lized software system is untenable# A technique 
for retrieval program generation was devised whe¬ 
rein a custom program, tailored to the specific 
queries submitted, and to the specific data format 
involved, is created on each processing cycle. 
Despite the diversity of data parameters en¬ 
compassed, four major elements of similarity have 
been identified. All participants wish to access 
data by taxa name, by geographical or political 
location; by catalog number; and in certain cases 
by stratigraphic designators. The data base 
comprised of specimen and bibliographic data is 
maintained in a dynamic phylogenetic sequence and 
is accessed by four indices: phylogenetic, geo¬ 
political-, catalog number (intervals) and stra¬ 
tigraphic, Other basic tenets of the system 
include: 
1. All consistently formated data is searchable. 
2. No abbreviation or user coding is imposed 
by the system. The computer translates 
natural language input into code and re¬ 
translates output. 
3. User training is minimized. 
4. A machine independent language is used (to 
enable exchange of software as well as of 
dat a) . 
The system is now being improved and implemented 
by the Information Systems Division, Smithsonian Insti 
tution, Washington, D.C. 
KEYWORD ELEMENTS: 
Retrieval program generation,dynamic phylogeny,hu¬ 
man factors of query,machine independent query,generali 
zed data collection,locational heirarchy,common denomina 
tor indices,synonomy rectification,heirarchical indices, 
biological numericlature, query batch optimization. 
3. "PREPARACION Y MANIPULACION DE CLAVES SISTEMAT1CAS 
UTILIZANDO COMPUTADORAS DE TIEMPO COMPARTIDO" 
James A. Peters 
Smithsonian Institution 
Washington, D.C. 
Se ha construldo una clave para identificacion de los gene- 
ros de culebras neotrdpicas utilizando una computadora de tiempo 
compartido y el lenguaje llamado "BASIC". La simplicidad de es- 
te lenguaje ha permitido a la computadora usar solamente catorce 
caracterlsticas, las cuales proveen toda la informacidn necesaria 
para las identificaciones. Se puaden inclulr otras caracterlsti¬ 
cas dentro del programa cuando hay un genera unico, y la computa¬ 
dora escribe la informacidn adicional en ingles, con los nombres 
genericos pertinentes. Cuando hay un genero nuevo o uno no inclul 
do anteriormente en la clave, se puede incluir con gran facilidad. 
Necesita nada mas que desafiar a la computadora con las catorce 
caracterlsticas del nuevo genero, porque esta aprovecha un norr.bre 
generico antiguo. Se puede comparar esta con el genero nuevo, y 
distinguir entre los das para facilitar la introduccidn del nuevo. 
Todos las generos conocidos hasta el momento de Guatemala hasta 
Argentina estan en la clave. Otros especialistas que quieran usar 
la clave, pueden pedir directamente una copia que sea aceptada por 
la computadora, o, si no tienen una computadora de tiempo comparti 
do, alias pueden mandar una lista de los ejemplares con las cator¬ 
ce caracterlsticas para desafiar a nuestra computadora. 
