Stochastic Models in Reliability Engineering

(Advanced Research in Reliability and System

Assurance Engineering) 1st Edition Lirong Cui
(Editor) pdf download
https://textbookfull.com/product/stochastic-models-in-
reliability-engineering-advanced-research-in-reliability-and-
system-assurance-engineering-1st-edition-lirong-cui-editor/

Download more ebook from https://textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Advances in Reliability and System Engineering 1st

Edition Mangey Ram

https://textbookfull.com/product/advances-in-reliability-and-
system-engineering-1st-edition-mangey-ram/

Applied Systems Analysis: Science and Art of Solving

Real-Life Problems (Advanced Research in Reliability
and System Assurance Engineering) 1st Edition F. P.
Tarasenko
https://textbookfull.com/product/applied-systems-analysis-
science-and-art-of-solving-real-life-problems-advanced-research-
in-reliability-and-system-assurance-engineering-1st-edition-f-p-
tarasenko/

Reliability Engineering Probabilistic Models and

Maintenance Methods Second Edition Nachlas

https://textbookfull.com/product/reliability-engineering-
probabilistic-models-and-maintenance-methods-second-edition-
nachlas/

Reliability engineering. Probabilistic models and

maintenance methods Second Edition Nachlas

https://textbookfull.com/product/reliability-engineering-
probabilistic-models-and-maintenance-methods-second-edition-
nachlas-2/
Site Reliability Engineering 1st Edition Betsy Beyer

https://textbookfull.com/product/site-reliability-
engineering-1st-edition-betsy-beyer/

Reliability Engineering 3rd Edition Elsayed A. Elsayed

https://textbookfull.com/product/reliability-engineering-3rd-
edition-elsayed-a-elsayed/

Reliability engineering theory and practice 8th Edition

Birolini A

https://textbookfull.com/product/reliability-engineering-theory-
and-practice-8th-edition-birolini-a/

Stochastic Models in Reliability Network Security and

System Safety Essays Dedicated to Professor Jinhua Cao
on the Occasion of His 80th Birthday Quan-Lin Li

https://textbookfull.com/product/stochastic-models-in-
reliability-network-security-and-system-safety-essays-dedicated-
to-professor-jinhua-cao-on-the-occasion-of-his-80th-birthday-
quan-lin-li/

Theory and Practice of Quality and Reliability

Engineering in Asia Industry 1st Edition Cher Ming Tan

https://textbookfull.com/product/theory-and-practice-of-quality-
and-reliability-engineering-in-asia-industry-1st-edition-cher-
ming-tan/
 Stochastic Models in
Reliability Engineering
 Advanced Research in Reliability and System
Assurance Engineering
Series Editor: Mangey Ram, Professor, Graphic Era University,
Uttarakhand, India

Modeling and Simulation Based Analysis in Reliability

Engineering
Edited by Mangey Ram

Reliability Engineering
Theory and Applications
Edited by Ilia Vonta and Mangey Ram

System Reliability Management

Solutions and Technologies
Edited by Adarsh Anand and Mangey Ram

Reliability Engineering
Methods and Applications
Edited by Mangey Ram

Reliability Management and Engineering

Challenges and Future Trends
Edited by Harish Garg and Mangey Ram

Applied Systems Analysis

Science and Art of Solving Real-Life Problems
F. P. Tarasenko

Stochastic Models in Reliability Engineering

Lirong Cui, Ilia Frenkel, and Anatoly Lisnianski

For more information about this series, please visit: https://www.crcpress.

com/Advanced-Research-in-Reliability-and-System-Assura nce-
Engineering/book-series/CRCARRSAE
 Stochastic Models in
Reliability Engineering

Edited by
Lirong Cui, Ilia Frenkel, and Anatoly Lisnianski
First edition published 2021
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742

and by CRC Press

2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

© 2021 Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, LLC

Reasonable efforts have been made to publish reliable data and information, but the author and publisher can-
not assume responsibility for the validity of all materials or the consequences of their use. The authors and
publishers have attempted to trace the copyright holders of all material reproduced in this publication and
apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright
material has not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmit-
ted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying, microflming, and recording, or in any information storage or retrieval system, with-
out written permission from the publishers.

For permission to photocopy or use material electronically from this work, access www.copyright.com or
contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400.
For works that are not available on CCC please contact [email protected]

Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only
for identifcation and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Names: Cui, Lirong, 1960- editor. | Frenkel, Ilia, 1950- editor. |

Lisnianski, Anatoly, editor.
Title: Stochastic models in reliability engineering / edited by Lirong Cui,
Ilia Frenkel and Anatoly Lisnianski.
Description: First edition. | Boca Raton, FL : CRC Press/Taylor & Francis
Group, 2020. | Includes index.
Identifers: LCCN 2020010183 (print) | LCCN 2020010184 (ebook) | ISBN
9780367345853 (hardback ; acid-free paper) | ISBN 9780429331527 (ebook)
Subjects: LCSH: Reliability (Engineering)--Statistical methods. |
Stochastic analysis.
Classifcation: LCC TA169 .S759 2020 (print) | LCC TA169 (ebook) | DDC
620/.004520151922--dc23
LC record available at https://lccn.loc.gov/2020010183
LC ebook record available at https://lccn.loc.gov/2020010184

ISBN: 9780367345853 (hbk)

ISBN: 9780429331527 (ebk)

Typeset in Times
by Deanta Global Publishing Services, Chennai, India

MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks
does not warrant the accuracy of the text or exercises in this book. This book’s use or discussion
of MATLAB® software or related products does not constitute endorsement or sponsorship by The
MathWorks of a particular pedagogical approach or particular use of the MATLAB® software.
Contents
Preface.......................................................................................................................ix
Editors.......................................................................................................................xi
Contributors ........................................................................................................... xiii

Chapter 1 Reliability Analysis of a Pseudo Working Markov

Repairable System................................................................................1
Bei Wu and Lirong Cui

Chapter 2 System Reliability Assessment with Multivariate Dependence

Models ................................................................................................ 19
Rong Pan and Guanqi Fang

Chapter 3 Reliability Modelling of Multi-Phased Linear Consecutively

Connected Systems ............................................................................ 41
Kaiye Gao, Xiangbin Yan, Hui Xiao, and Rui Peng

Chapter 4 A Method for Complex Multi-State Systems Reliability

Analysis Based on Compression Inference Algorithm and
Bayesian Network............................................................................... 49
Xiaohu Zheng, Wen Yao, Yingchun Xu, and Xiaoqian Chen

Chapter 5 Reliability Analysis of Demand-Based Warm Standby System

with Multi-State Common Bus .......................................................... 67
Yi Ding, Heping Jia, and Rui Peng

Chapter 6 An Upside-Down Bathtub-Shaped Failure Rate Model Using a

DUS Transformation of Lomax Distribution ..................................... 81
K.S.Deepthi and V.M.Chacko

Chapter 7 Reliability Analysis of Multi-State Systems with Dependent

Failures Based on Copula................................................................. 101
Xujie Jia, Xueying Song, and Gang Li

v
vi Contents

Chapter 8 Modelling and Inference for Special Types of Semi-Markov

Processes .......................................................................................... 113
Vlad Stefan Barbu, Alex Karagrigoriou, and Andreas Makrides

Chapter 9 Weighted Multi-Attribute Acceptance Sampling Plans ................... 125

Xian Zhao, Congshan Wu, Leping Sun, and Yu Fan

Chapter 10 Reliability Assessment for Systems Suffering Common

Cause Failure Based on Bayesian Networks and Proportional
Hazards Model ................................................................................. 135
Yan-Feng Li, Yang Liu, Hong-Zhong Huang, Tudi Huang, and
Jinhua Mi

Chapter 11 Early Warning Strategy of Sparse Failures for Highly Reliable

Products Based on the Bayesian Method ......................................... 151
Shirong Zhou, Yincai Tang, Ancha Xu, and Yongqiang Lian

Chapter 12 Fault Detection and Prognostics of Aero Engine by Sensor

Data Analytics.................................................................................. 163
Haotian Liu and Xi Zhang

Chapter 13 Stochastic Modelling of Opportunistic Maintenance for

Series Systems with Degrading Components .................................. 183
Vasilis Koutras, Sonia Malefaki, and Agapios N. Platis

Chapter 14 On Censored and Truncated Data in Survival Analysis and

Reliability Models ............................................................................ 199
Catherine Huber, Valentin Solev, and Filia Vonta

Chapter 15 Analysis of Node Resilience Measures for Network Systems ......... 213
Chao Zhang, Xin Xu, and Hongyan Dui

Chapter 16 Reliability Analysis of General Purpose Parts for Special

Vehicles Based on Durability Testing Technology .......................... 227
Guang-Jun Jiang, Nan Zhang,
Qing-Chao Zhang, and Ying Wang
Contents vii

Chapter 17 State of Health Prognostics of Lithium-Ion Batteries ...................... 245

Huixing Meng and Yan-Fu Li

Chapter 18 Life Prediction of Device Based on Material’s Micro-Structure

Evolution by Means of Computational Materials Science ............... 263
Chunyu Teng, Cong Lin, Zhanyong Ren, Yun Fu, and
Xichang Wang

Chapter 19 Low-Cycle Fatigue Damage Assessment of Turbine Blades

Using a Substructure-Based Reliability Approach .......................... 281
Haifeng Gao, Enrico Zio, Anjenq Wang, and Guangchen Bai

Chapter 20 Phased-Mission Modelling of Physical Layer Reliability for

Smart Homes.................................................................................... 317
Liudong Xing, Guilin Zhao, and Yisha Xiang

Chapter 21 Comparative Reliability Analysis of Different Traction Drive

Topologies for a Search-and-Rescue Helicopter .............................. 331
Igor Bolvashenkov, Jörg Kammermann, Zeng Wenbin, Ilia
Frenkel, and Hans-Georg Herzog

Chapter 22 Reliability and Fault Tolerance Assessment of Different

Operation Modes of Air Conditioning Systems for Chemical
Laboratories...................................................................................... 355
Lev Khvatskin, Ilia Frenkel, and Anatoly Lisnianski

Chapter 23 Dependability Analysis of Ship Propulsion Systems ....................... 373

Thomas Markopoulos and Agapios N. Platis

Chapter 24 Application of Markov Reward Processes to Reliability, Safety,

Performance Analysis of Multi-State Systems with Internal and
External Testing ............................................................................... 391
Valentina S. Viktorova, Nikolay V. Lubkov,
and Armen S. Stepanyants

Chapter 25 Multi-Objective Maintenance Optimization of Complex

Systems Based on Redundancy-Cost Importance............................407
Chenyang Ma, Wei Wang, and Zhiqiang Cai
viii Contents

Chapter 26 Which Replacement Maintenance Policy Is Better for Multi-

State Systems?: Policy T or Policy N?.............................................. 419
Wenjie Dong, Sifeng Liu, and Suk Joo Bae

Chapter 27 Design of Multi-Stress Accelerated Life Testing Plans Based

on D-Optimal Experimental Design ................................................ 431
Xiangxiang Zhang, Jun Yang, and Xuefeng Kong

Chapter 28 An Extended Optimal Replacement Policy for a Simple

Repairable Modelling....................................................................... 443
Junyuan Wang, Jimin Ye, and Ning Ma
Index...................................................................................................................... 461
Preface
In this book, the 28 chapters present recent advances in reliability theory and its appli-
cations, life science, and operation management. All chapters in the book are writ-
ten by leading researchers and practitioners in their respective felds of expertise and
present a number of innovative methods, approaches, and solutions not covered before
in the literature. This book will be helpful to post-graduate and doctoral students,
researchers, practitioners, and industrial managers interested in corresponding felds.
Many of the chapters were presented as papers at the Third International
Symposium on Stochastic Models in Reliability Engineering, Life Sciences, and
Operations Management (SMRLO-2019) held in Beijing, China, from May 28 to
May 31, 2019. The objective of the SMRLO-2019 conference was to provide a forum
for scholars and practitioners to share the results of latest research efforts, visions
for the future, and experiences in the stochastic modelling in reliability engineering,
life sciences, and operations management. The conference was highly successful,
receiving more than 120 papers from 13 countries and regions. Some of these have
been selected for presentation in this book as well as in three international journals.
The SMRLO-2019 conference was hosted by Industrial Engineering Committee,
Chinese Society of Optimization, Overall Planning, and Economical Mathematics
and School of Economics and Management, Beijing Institute of Technology, which
was the frst time that the SMRLO conference was held in China. The confer-
ence was initiated by several Israeli professors. It was frst held in Beer Sheva,
Israel, February 15–17, 2005, called the ‘International Symposium on Stochastic
Models in Reliability, Safety, Security, and Logistics’. In 2010, the conference was
renamed ‘The First International Symposium on Stochastic Models in Reliability
Engineering, Life Sciences, and Operations Management’ and held in Beer Sheva,
Israel, February 8–11, 2010. The second symposium was also held in Beer Sheva,
Israel, February 15–18, 2016.
We would like to thank all members of the programme committee for their pro-
moting the conference, thank all the authors for their contributions and reviewers
for their tireless efforts to improve the contributed papers and chapters. Also, we
should acknowledge the conference organizers for their effort in organizing this
conference. We appreciate the fnancial support from the National Natural Science
Foundation of China. Finally, we would like to thank CRC/Taylor and Francis
Group for help with this publication. It was indeed our pleasure to work with the
CRC editorial team.

Lirong Cui
(Beijing Institute of Technology, China)
Ilia Frenkel
(Shamoon College of Engineering, Israel)
Anatoly Lisnianski
(The Israel Electric Corporation, Israel)

ix
x Preface

MATLAB® is a registered trademark of The MathWorks, Inc. For product infor-

mation, please contact:

The MathWorks, Inc.

3 Apple Hill Drive
Natick, MA 01760-2098 USA
Tel: 508 647 7000
Fax: 508-647-7001
E-mail: [email protected]
Web: www.mathworks.com
Editors
Lirong Cui is a professor in the School of Management and Economics at Beijing
Institute of Technology. He received his PhD degree in probability and statistics
from the University of Wales (Swansea, UK), in 1994. He has worked on quality
and reliability-related problems since 1986 and published more than 120 papers and
technical reports. He has served as an associate editor for several journals. In 2005,
he was awarded the New Century Excellent Talents in China. His recent research
interests are in stochastic modeling, Hawkes processes, reliability engineering, opti-
mization, operations research, and applications of probability, and statistics in vari-
ous felds.

Ilia Frenkel is a researcher in the Center for Reliability and Risk Management,
SCE–Shamoon College of Engineering, Beer Sheva, Israel. He obtained his MSc
degree in Applied Mathematics from Voronezh State University, Russia, and PhD
degree in Operational Research and Computer Science from Institute of Economy,
Ukrainian Academy of Science, formerly the USSR. He has more than 40 years
of academic experience and teaching in universities and institutions in Russia and
Israel. From 1988 to 1991 he worked as department chair and associate professor in
the applied mathematics department and at computers department, Volgograd Civil
Engineering Institute, Russia. From 2001 he served as senior lecturer in the indus-
trial engineering and management department and from 2005 as Chair of the Center
for Reliability and Risk Management, SCE–Shamoon College of Engineering, Beer
Sheva, Israel. He retired in 2018. He has specialized in applied statistics and reliabil-
ity with application to preventive maintenance. He published two books and more
than 50 scientifc articles and book chapters in the felds of reliability, applied sta-
tistics, and production and operation management. He is an editor and member of
several editorial boards of scientifc and professional journals.

Anatoly Lisnianski is a senior engineering expert in the reliability department of

the Israel Electric Corporation Ltd., Israel. Lisnianski received his MSc degree in
Electrical Engineering from the University of Information Technologies, Precision
Mechanics, and Optics, Sankt-Petersburg, Russia, and his PhD degree in Reliability
in 1984 from the Federal Scientifc and Production Center ‘Aurora’ in St. Petersburg.
From 1975 to 1989, he was a researcher at this center. In 1991 he began working in
the reliability department of the Israel Electric Corporation as an engineering expert.
He specializes in reliability assessment and optimization for complex technical sys-
tems. From 2005 he served as scientifc supervisor of the Center for Reliability and
Risk Management, SCE–Shamoon College of Engineering, Beer Sheva, Israel. He is
the author and co-author of more than 120 journal papers, three books, several book
chapters, and inventions in the feld of reliability and applied statistics. He is a senior
member of IEEE and a member of the Israel Statistical Association.

xi
Contributors
Suk Joo Bae Lirong Cui
Department of Industrial School of Management and Economics
Engineering Beijing Institute of Technology
Hanyang University Beijing, China
Seoul, South Korea
K.S. Deepthi
Guangchen Bai Department of Statistics
School of Energy and Power University of Calicut
Engineering Kerala, India
Beijing University of Aeronautics and
Astronautics Yi Ding
Beijing, China College of Electrical Engineering
Zhejiang University
Vlad Stefan Barbu Hangzhou, China
Laboratoire de Mathématiques Raphaël
Salem Wenjie Dong
Université de Rouen College of Economics and Management
Saint-Étienne-du-Rouvray, France Nanjing University of Aeronautics and
Astronautics
Igor Bolvashenkov Nanjing, China
Institute of Energy Conversion
Technology Hongyan Dui
Technical University of Munich School of Management Engineering
Munich, Germany Zhengzhou University
Zhengzhou, China
Zhiqiang Cai
School of Mechanical Engineering Yu Fan
Northwestern Polytechnical School of Management and Economics
University Beijing Institute of Technology
Xi’an, China Beijing, China

V.M. Chacko Guanqi Fang

Department of Statistics School of Computing, Informatics, and
University of Calicut Decision Systems Engineering
Kerala, India Arizona State University
Tempe, AZ
Xiaoqian Chen
National Innovation Institute of Defense Ilia Frenkel
Technology Center for Reliability and Risk
Chinese Academy of Military Management
Science SCE–Shamoon College of Engineering
Beijing, China Beer Sheva, Israel

xiii
xiv Contributors

Yun Fu Xujie Jia

AVIC China Aero-Polytechnology College of Science
Establishment Minzu University of China
Beijng, China Beijing, China

Haifeng Gao Guang-Jun Jiang

School of Mechanical Engineering School of Mechanical Engineering
Shanghai Jiao Tong University Inner Mongolia University of
Shanghai, China Technology
Hohhot, China
Kaiye Gao
School of Economics & Jörg Kammermann
Management Institute of Energy Conversion
Beijing University of Information Technology
Science & Technology Technical University of Munich
Beijing, China Munich, Germany

Hans-Georg Herzog Alex Karagrigoriou

Institute of Energy Conversion School of Sciences
Technology Department of Statistics and Actuarial-
Technical University of Munich Financial Mathematics
Munich, Germany University of the Aegean
Samos, Greece
Hong-Zhong Huang
School of Mechanical and Electrical Lev Khvatskin
Engineering Center for Reliability and Risk
University of Electronic Science and Management
Technology of China SCE–Shamoon College of Engineering
Chengdu, China Beer Sheva, Israel

Tudi Huang Xuefeng Kong

School of Mechanical and Electrical School of Reliability and Systems
Engineering Engineering
University of Electronic Science and Beihang University
Technology of China Beijing, China
Chengdu, China
Vasilis Koutras
Catherine Huber Department of Financial and
MAP5 CNRS Management Engineering
Université de Paris University of the Aegean
Paris, France Chios, Greece

Heping Jia Gang Li

College of Electrical Engineering College of Science
Zhejiang University Minzu University of China
Hangzhou, China Beijing, China
Contributors xv

Yan-Feng Li Nikolay V. Lubkov

School of Mechanical and Electrical Laboratory of Technical Diagnostics
Engineering and Fault Tolerance
University of Electronic Science and V.A. Trapeznikov Institute of Control
Technology of China Sciences RAS
Chengdu, China Moscow, Russia

Yan-Fu Li Chenyang Ma
Department of Industrial Engineering School of Mechanical Engineering
Tsinghua University Northwestern Polytechnical University
Beijing, China Xi’an, China

Yongqiang Lian Ning Ma

School of Statistics School of Mathematics and Statistics
East China Normal University Xidian University
Shanghai, China Xi’an, China

Cong Lin Andreas Makrides

AVIC China Aero-Polytechnology Department of Mathematics and
Establishment Statistics
Beijing, China University of Central Lancashire
Larnaka, Cyprus
Anatoly Lisnianski
The Reliability Department Sonia Malefaki
The Israel Electric Corporation Department of Mechanical Engineering
Haifa, Israel and Aeronautics
University of Patras
Haotian Liu Rio Patras, Greece
Department of Industrial Engineering
and Management Thomas Markopoulos
Peking University Department of Financial and
Beijing, China Management Engineering
University of the Aegean
Sifeng Liu Chios, Greece
College of Economics and
Management Huixing Meng
Nanjing University of Aeronautics and Department of Industrial Engineering
Astronautics Tsinghua University
Nanjing, China Beijing, China

Yang Liu Jinhua Mi

School of Mechanical and Electrical School of Mechanical and Electrical
Engineering Engineering
University of Electronic Science and University of Electronic Science and
Technology of China Technology of China
Chengdu, China Chengdu, China
xvi Contributors

Rong Pan Yincai Tang

School of Computing, Informatics, and School of Statistics
Decision Systems Engineering East China Normal University
Arizona State University Shanghai, China
Tempe, AZ
Chunyu Teng
Rui Peng AVIC China Aero-Polytechnology
School of Economics and Management Establishment
Beijing University of Technology Beijing, China
Beijing, China
Valentina S. Viktorova
Laboratory of Technical Diagnostics
Agapios N. Platis and Fault Tolerance
Department of Financial and V.A. Trapeznikov Institute of Control
Management Engineering Sciences RAS
University of the Aegean Moscow, Russia
Chios, Greece
Filia Vonta
Zhanyong Ren Department of Mathematics
AVIC China Aero-Polytechnology National Technical University of Athens
Establishment Athens, Greece
Beijing, China
Anjenq Wang
Valentin Solev School of Mechanical Engineering
Laboratory of Statistical Methods Shanghai Jiao Tong University
Steklov Institute of Mathematics at St. Shanghai, China
Petersburg and
St. Petersburg, Russia
Department of Energy
Politecnico di Milano
Xueying Song Milan, Italy
College of Science
Minzu University of China Junyuan Wang
Beijing, China School of Mathematics and Statistics
Xidian University
Armen S. Stepanyants Xi’an, China
Laboratory of Technical Diagnostics
and Fault Tolerance Wei Wang
V.A. Trapeznikov Institute of Control School of Mechanical Engineering
Sciences RAS Northwestern Polytechnical University
Moscow, Russia Xi’an, China

Leping Sun Xichang Wang

School of Management and Economics AVIC China Aero-Polytechnology
Beijing Institute of Technology Establishment
Beijing, China Beijing, China
Contributors xvii

Ying Wang Yingchun Xu

School of Mechanical Engineering College of Aerospace Engineering
Inner Mongolia University of National University of Defense
Technology Technology
Hohhot, China Changsha, China

Zeng Wenbin Xiangbin Yan

School of Mechanical and Aerospace Donlinks School of Economics &
Engineering Management,
Jilin University University of Science & Technology
Changchun, China Beijing
Beijing, China
Bei Wu
School of Management and Economics Jun Yang
Beijing Institute of Technology School of Reliability and Systems
Beijing, China Engineering
Beihang University
Congshan Wu Beijing, China
School of Management and Economics
Beijing Institute of Technology Wen Yao
Beijing, China National Innovation Institute of Defense
Technology
Yisha Xiang Chinese Academy of Military
Department of Industrial Engineering Science
Texas Tech University Beijing, China
Lubbock, TX
Jimin Ye
Hui Xiao School of Mathematics and Statistics
School of Statistics, Xidian University
Southwestern University of Finance and Xi’an, China
Economics
Chengdu, China Chao Zhang
School of Automation Science and
Liudong Xing Electrical Engineering
Department of Electrical and Computer Beihang University
Engineerging Beijing, China
University of Massachusetts
Dartmouth, MA Nan Zhang
School of Mechanical Engineering
Ancha Xu Inner Mongolia University of
Department of Statistics Technology
Zhejiang Gongshang University Hohhot, China
Hangzhou, China
Qing-Chao Zhang
Xin Xu School of Mechanical Engineering
School of Management Engineering Inner Mongolia University of
Zhengzhou University Technology
Zhengzhou, China Hohhot, China
xviii Contributors

Xi Zhang Shirong Zhou

Department of Industrial Engineering School of Statistics
and Management East China Normal University
Peking University Shanghai, China
Beijing, China
Enrico Zio
Xiangxiang Zhang Department of Energy
School of Reliability and Systems Politecnico di Milano
Engineering Milan, Italy
Beihang University
Beijing, China and
MINES Paris Tech
Guilin Zhao
PSL Research University, CRC
Department of Electrical and Computer
Sophia Antipolis, France
Engineerging
University of Massachusetts and
Dartmouth, MA
Department of Nuclear Engineering,
Xian Zhao College of Engineering
School of Management and Economics Kyung Hee University
Beijing Institute of Technology Seoul, Korea
Beijing, China

Xiaohu Zheng
College of Aerospace Engineering
National University of Defense
Technology
Changsha, China
 1 Reliability Analysis of a
Pseudo Working Markov
Repairable System
Bei Wu and Lirong Cui

CONTENTS
1.1 Introduction ...................................................................................................... 1
1.2 Basic Assumptions............................................................................................3
1.3 Reliability Indexes ............................................................................................6
1.3.1 Case of Constant τ ................................................................................ 6
1.3.1.1 Time to First Failure ..............................................................6
1.3.1.2 Point-Wise and Interval Availabilities ...................................7
1.3.2 Case of Random τ............................................................................... 11
1.4 Numerical Examples and Special Cases ........................................................ 13
1.5 Conclusions..................................................................................................... 16
References................................................................................................................ 17

1.1 INTRODUCTION
A reliability analysis of a Markov repairable system has been a hot topic for some time
and various models based on Markov processes have been developed. For example,
Liu et al. (2015) proposed a cold standby repairable system with working vacations
and vacation interruption following a Markovian arrival process. Montoro-Cazorla
and Pérez-Ocón (2014) studied a reliability system under different types of shock
governed by a Markovian arrival process and maintenance policy. Du et al. (2016)
developed a degradation model for repairable systems based on a continuous-time
Markov process with multiple discrete states.
These models gave clear defnitions of failure and working states, but in prac-
tice, there are situations considering time omission problems where systems can be
regarded as operating in failure states under some restrictions. A new model where
failure can be thought of as an operation time during repair time was frstly proposed
in Zheng et al. (2006). It was extended to the series Markov repairable system with
neglected or delayed failures by Bao and Cui (2010), and to the aggregated Markov
repairable system with repair time omission which considers that one complex sys-
tem may have hundreds of states by Liu et al. (2013a).
While under some circumstances, system operation can be thought of as a failure
time in working time. It is commonly seen in the power station system, where only
after the power station recovers from failure and runs normally for a period of time,

1
2 Stochastic Models in Reliability Engineering

the users think the power station has been well repaired. Yi and Cui (2017) modelled
discrete repairable degradation systems with working time omission by a homoge-
neous discrete-time second-order semi-Markov chain. Zhang et al. (2017) studied
the performance of a single-unit repairable system consisting of two states (up and
down) with working and repair time omission by an alternative renewal process.
This chapter is an extension of Zheng et al. (2006) and Zhang et al. (2017), through
generalizing exponential distributions to Phase-type distributions. In the present
chapter, an irreducible homogenous continuous-time Markov process {X (t ), t ³ 0}
is proposed to model the evolution of a system, in which not enough long operation
time will be regarded as a failure time. The formulas for reliability indexes of the
system under the proposed new model are derived by exploiting the theory of aggre-
gated stochastic processes.
The theory of aggregated stochastic processes was introduced by Colquhoun
and Hawkes (1982) to model the behaviour of ion-channels. Then many researchers
greatly enriched this theory, such as Jalali and Hawkes (1992a, 1992b), Merlushkin
and Hawkes (1997) and Ball (1987, 1999). The theory of aggregated stochastic pro-
cesses has been widely used in the reliability feld theses years, and various models
have been developed to describe repairable systems, such as Cui et al. (2012, 2014),
Liu et al. (2013b), Li et al. (2016), Wen et al. (2016) and Du et al. (2018). This chapter
applies the theory of aggregated stochastic processes to derive the formulas for the
distribution of time to frst failure, point-wise, and interval availabilities of the new
proposed repairable system.
Various performance metric indexes have been studied by many researchers
(e.g., Levitin et al., 2014, Liu et al., 2017, Hao and Yang, 2018, and Yu et al., 2018).
Availabilities are considered in repairable systems such as the point-wise availability
and interval availability, which is the probability that the system will be able to work
within the tolerances at a given instant time or a time interval respectively (Barlow
and Proschan, 1965). Cui and Xie (2001) studied the availability of the periodically
inspected system under the random walk model, and an extended case combining
random repair and replacement time was discussed by Cui and Xie (2005). Qiu et
al. (2017) derived the formulas for instantaneous and steady-state availabilities for
a competing-risk system undergoing periodic inspections. Many reliability indexes
including the distribution of the time to frst failure, mean time to frst failure, point-
wise, steady, and interval availabilities are discussed in this chapter.
Engines of motor vehicles operate within a specifc speed range which is limited
by the idle speed and the maximum speed (Reif, 2014). They do not have a constant
torque and power in such a speed range, but they have an optimal elastic speed range
between the maximum torque and the maximum power. Hence, a vehicle cannot
start running directly when the engine stays in its stationary state. If the vehicle has
not been running for many hours before starting, the engine will usually run at a low
rpm (600–1000 rpm), which is called cold start. If the engine operates without any
loads and the vehicle is not in motion, this state is called idling.
The effciency of the vehicle engine is not a constant number, but changes with
rpm. The transmission aims to keep the engine in its peak effciency. When shifting
up different higher gears and accelerating, the vehicle engine goes to a high rpm since
more power is required. For normally driving, the operating range for most engines
Reliability Analysis of Pseudo Markov System 3

FIGURE 1.1 A possible state transition route of the vehicle engine.

is in the range of 1500–3000 rpm. 2000 rpm is an ideal rpm for constant speed driv-
ing. Hence, on the basis of the engine speed, engine states can be discretized into a
state space which is defned by S = {0,1, , 5} where states 1 to 5 denote 650 rpm,
1500 rpm, 2000 rpm, 2500 rpm, 3000 rpm respectively. State 0 represents 0 rpm
which means the vehicle engine is at rest. The vehicle is stationary but the engine is
running when it stays at state 1 which is the idle state. The vehicle usually stays in
its idle state for 30 seconds to 60 seconds. If the engine is fameout and goes to state
0 when it stays at state 1 before transferring to other states, the vehicle has no actual
displacement during this period, which is called the pseudo working time. A possible
state transition route of the vehicle engine is shown in Figure 1.1.
The rest of the content is organized as follows. Basic assumptions of our model
are given in Section 1.2. Reliability indexes such as the distributions of the time to
frst failure, mean time to frst failure, point-wise availabilities, steady availabilities
and interval availabilities under two cases when τ is a constant and a random variable
are discussed respectively in Section 1.3. After obtaining the closed form formu-
las for reliability indexes, some numerical examples are presented to illustrate the
results in Section 1.4. Finally, in Section 1.5, conclusions and future research work
are provided.

1.2 BASIC ASSUMPTIONS

Suppose that a multi-state repairable system can be described by an irreducible
homogenous continuous-time Markov process {X (t ), t ³ 0} with a fnite state space
S = {0,1, , n}. These discrete states can be divided into two subsets, a subset of
working states W, and a subset of failure states F, i.e., S = W  F and W  F = f .
Hence, for the vehicle engine system, states 1 to 5 can be aggregated into the work-
ing subset W, and state 0 belongs to the failure subset F.
The infnitesimal generator of the process {X (t ), t ³ 0} is Q which can be divided
into four blocks in terms of the two subsets, i.e.,

æ QWW QWF ö
Q=ç ÷.
è QFW QFF ø
4 Stochastic Models in Reliability Engineering

It is a standard result that:

P(t ) = exp(Qt),

where P(t) whose ijth element is Pij (t ) = P(state j at time t state i at time 0) , denotes
the probability of the repairable system being in each state of S at time instant t,
given the initial probability vector. The initial probability is p 0 = (p W , p F ).
We list some basic notations and results that shall be used in the following deri-
vation. Two important performance measures were introduced by Colquhoun and
Hawkes (1982) as follows.
pij ,WW (t ) = P(system remains within W throughout time 0 to tiime t,

and is in state j at time t in state i at time 0), i,, j Î W .

PWW(t) is defned as a |W| × |W| matrix whose ijth element is pij ,WW (t ). We have

PWW (t ) = exp(QWW t). (1.1)

The Laplace transform of Equation (1.1) is:

*
PWW (s) = (sI - QWW )-1,

where I is a unit matrix with proper dimensions, and s is the complex argument. GWF(t)
is a |W| × |F| matrix whose ijth element is gij(t), which is defned as

gij (t ) = lim éë P ( system stays in W from time 0 to time t, andd leave

Dt ®0

D ù,
W for state j between t and t + Dt in state i at time 0 /Dt
û )
i Î W , j Î F.
We can fnd that:
GWF (t ) = PWW (t )QWF . (1.2)

Taking Laplace transform on both sides of Equation (1.2) yields:

* *
GWF (s) = PWW (s)QWF = (sI - QWW )-1 QWF . (1.3)

Setting s = 0 in Equation (1.3), we have:

* -1
GWF := GWF (0) = - QWW QWF . (1.4)

GWF gives the transition probabilities from one of the states in W to one of the states
in F, allowing for any number of transitions within the states in W before the system
eventually leaves for F.
Reliability Analysis of Pseudo Markov System 5

Hawkes et al. (1990) introduced two new indexes which were derived from divid-
*
ing the Laplace transform of PWW(t) into two parts: SWW (s) which describes a short
*
sojourn in W (smaller than τ) and LWW (s) which describes a long sojourn in W (not
smaller than τ). Their expressions are

ò
*
SWW (s)(sI - QWW ) = e -st exp(QWW t )dt
-1

0 (1.5)

{ (
= I - exp - ( sI - QWW )t )} (sII - Q
WW )-1,

ò
L*WW (s) ( sI - QWW ) = e -st exp ( QWW t ) dt
-1

t (1.6)

(
= exp - ( sI - QWW )t ) ( sI - Q )
WW
-1
.

In the present chapter, system operation can be thought of as in a failure time when
the system stays in W but fails before the length of sojourn time in W reaches a given
value τ. These short sojourns in W (less than τ) are defned as pseudo working time,
and long sojourns in W (not less than τ) are defned as effective working time. For
the vehicle engine system, τ is 30 seconds. The running time of the engine which is
less than τ is the pseudo working period because the vehicle does not run actually.
The following assumptions are presented to describe the new model:

(1) There are two statuses of the observation for the original system: working
and failure. The new model also has two subsets, which are state W* and
state F*. The state W* denotes the working period while state F* denotes
the failed period.
(2) When the original system is in the failure status, the new model is in the
failed state F*.
(3) When the original system starts to work but stays in the working period W
less than a specifed value τ, the new system is still in the failed state F*
during this pseudo working period.
(4) When the original system starts to work and stays in the working period W
not less than τ, the new system can be considered to stay in working state
W* from the time instant when the system enters the status W to the time
when the system breaks down.

A new stochastic process {Y (t ), t ³ 0} describes the running path for the new system,
with state space S * = {W * , F *}, which is defned as follows:

ïìW , if the new system is regarded as working at time t,

*
{Y (t ), t ³ 0} = í *
îï F , if the new system is regarded as failure at timee t.
6 Stochastic Models in Reliability Engineering

FIGURE 1.2 Possible sample paths of the original system and the new model.

Figure 1.2 shows some possible system evolution paths for both the original system
and the new model.

1.3 RELIABILITY INDEXES

In this section, we shall discuss several reliability indexes of systems under the cir-
cumstances when τ is a constant or a random variable respectively. The formulas of
distributions of the time to frst failure, mean time to frst failure, point-wise, and
interval availabilities are given.

1.3.1 CASE OF CONSTANT τ

1.3.1.1 Time to First Failure
Suppose that the system begins working at time 0, X1 describes the time to frst failure.
Failure here has two meanings: one is the system enters failure states after enough long
working time, the other is the sojourn time in working states does not exceed τ. Then
we can obtain the Laplace transform of the probability density function of X1 as follows.
If t < τ,
*
p W SWW (s)(sI - QWW )-11W
f X*1 (s) = , (1.7)
p W LWW GWF 1F + p W SWW (-QWW )-1 1W
if t ≥ τ,

p W L*WW (s)GWF
*
(s )1F
f X*1 (s) = , (1.8)
p W LWW GWF 1F + p W SWW (-QWW )-1 1W
where 1W = (1,,1)TW ´1, and T indicates transposition.
Reliability Analysis of Pseudo Markov System 7

LWW and SWW in Equations (1.7) and (1.8) are obtained by setting s = 0 in L*WW (s) and
*
SWW (s) , which depict the probabilities of system staying in W for not smaller and
smaller than τ, respectively. Equations (1.7) and (1.8) are obtained since when t < τ,
the system fails only if the situation where the sojourn time in working states does not
exceed τ occurs; when t ≥ τ occurs, the system goes through a long sojourn in W and
then transfers to F. Based on two well-known results: (1) the probability density func-
tion of the sum of random variables is the convolution of their individual probability
density functions; (2) the Laplace transform of the required probability density func-
tion is the product of the Laplace transforms of the individual probability density func-
tions, Equations (1.7) and (1.8) can be obtained. The denominators in Equations (1.7)
and (1.8) are used to make f x1 become a probability density function. Then, we have

p W exp(QWW t)QWF 1F I[t ³t ] + p W exp(QWW t)1W I[t <t ]

f X1 (t ) = , (1.9)
p W LWW GWF 1F + p W SWW (-QWW )-1 1W

where I[t <t ] is an indicator function which is 1 when t < τ.

The mean time to frst failure (MTTFF) is:

MTTFF = EX1 = tf X1 (t )dt

ò
0

¥ (1.10)

ò t éëp W exp(QWW t)QWF 1F I[t ³t ] + p W exp(QWW t )11W I[t <t ] ùû dt

= 0
.
p W LWW GWF 1F + p W SWW (-QWW )-1 1W

1.3.1.2 Point-Wise and Interval Availabilities

We list the point-wise and interval availabilities for the original Markov repairable
system here in order to make a comparison with them for the new model, which are
denoted as A X(t) and A X([a,b]).

AX (t ) = P( X (t ) Î W ) = p 0 exp(Qt)(1W , 0F )T , (1.11)

where 1W is a column vector of |W| ones and 0F is a column vector of |F| zeros.
Given a time interval [a,b],

AX ([ a, b]) = P { X (t ) Î W, for all t Î [ a, b]}

(1.12)
= p 0 exp(Qa) EW exp ( QWW (b - a) ) 1W ,

æ IW ´W ö
where EW = ç ÷ .
ç 0F ´W ÷
è ø S ´W
8 Stochastic Models in Reliability Engineering

For the new model, we discuss the point-wise availability frstly. The point-wise
availability under the new model of stochastic process {Y (t ), t ³ 0} is defned as
AY (t ) = P(Y (t) Î W * ).
In terms of whether there is transition from the failure states F to the working
states W, AY(t) can be divided into two cases.

a) When there is no transition from states F to states W,

if t < τ,

AY 01 (t ) = p W exp(QWWt )1W , (1.13)

if t ≥ τ,

AY 02 (t ) = p W exp(QWW t)1W . (1.14)

Combining Equations (1.13) and (1.14) yields,

AY 0 (t ) = p W exp(QWWt )1W I[t <t ] + p W exp(QWW t)1W I[t ³t ]. (1.15)

Equation (1.15) is obtained because if t < τ, the system has to work at

least for a length of τ to make sure that system is in state W* at time instant
t; if t > τ, since there is no transfer from states F to states W, the system has
to work from the beginning to time instant t.
b) When there is at least one transition from states F to states W,

min( t ,t )

AY 1 (t ) =
ò
0
p 0 exp ( Q(t - h) ) EF QFW exp(QWWt )1W dh

(1.16)
max( t ,t )

+
ò
t
p 0 exp ( Q(t - h)
h ) EF QFW exp(QWW h)1W dh,

æ 0W ´ F ö
where EF = ç ÷ .
ç IF´F ÷
è øS´F

The explanation of Equation (1.16) is that before entering the working state W*, the
system may transfer between W and F for any times and then transfer to W fnally,
which is denoted as p 0 exp(Q(t - h))EF QFW . Then, the system has to work at least
for a length of τ.
Combining Equations (1.15) and (1.16), we can fnd the point-wise availability
under the new model, AY(t), as follows:
Reliability Analysis of Pseudo Markov System 9

AY (t ) = AY 0 (t ) + AY1 (t )

= p W exp(QWWt )1W I[t <t ] + p W exp(QWW t)1W I[ t ³t ]

min( t ,t )

+
ò
0
p 0 exp(Q(t - h))EF QFW exp(QWWt )1W dh (1.17)

max( t ,t )

+
ò
t
p 0 exp(Q(t - h))E
EF QFW exp(QWW h)1W dh.

Especially, when t ≥ τ, taking Laplace transform to Equation (1.17) yields:

ò
AY* (s) = p W (sI - QWW )-11W + p 0 (sI - Q)-1 exp(-Qh)dhEF QFW exp(QWWt )1
0
) W

-1 -1
+ p 0 (sI - Q) EF QFW (sI - QWW ) 1W (1.18)
t

ò
- p 0 (sI - Q)-1 exp(-Qh) EF QFW exp(Q
0
p WW h)1W dh.

Then we can fnd the steady point-wise availability as follows,

AY = lim AY (t ) = lim sAY* (s)

t ®¥ s®0

= lim sp W (sI - QWW )-1 1W

s®0

ò exp(-Qh)dhE Q
-1
+ lim sp 0 (sI - Q) F FW exp(QWWt )1W (1.19)
s®0
0

+ lim sp 0 (sI - Q)-1 EF QFW (sI - QWW )-1 1W

s®0

s®0 ò
- lim s p 0 (sI - Q)-1 exp(-Qh) EF QFW exp(QWW h)1W dh.
0

Next, we discuss the interval availability under the new model. Given a time interval
[a,b], the interval availability under the new model of stochastic process {Y (t ), t ³ 0}
is defned as AY ([ a, b]) = P{Y (t ) = W * , for all t Î [ a, b]}. When calculating AY([a,b]),
three situations should be considered as follows.

a) If b < τ, without transitions, the system starts from W and keeps in W until
time b, but in order to meet the requirement of availability, the system has
to stay in W until time τ. Then, we have
10 Stochastic Models in Reliability Engineering

AY1 ([a, b]) = p W exp(QWWt )1W I{b<t }. (1.20)

b) If b ≥ τ, without transitions, the system starts from W and keeps working to

time b, then we can get

AY 2 ([a, b]) = p W exp(QWW b)1W I{b³t }. (1.21)

c) When there is at least one transition from states F to states W, in order to

calculate AY([a,b]), one important index which describes the last time instant
staying in state F, before time instant a when system is in state W proposed by
Cui et al. (2017), Ta := sup {u :u < a, Y (u) Î F , Y (a) Î W } is introduced here.
The Laplace transform of the probability density function of Ta is

¥ ¥

å å éëG
n n
fT*a (s) = p W é GWF
* *
(s)GFW (s) ùû + p F GFW
*
(s) * *
(s)GFW (s) ùû
ë WF
n=1 n=0 (1.22)

{
= p W éë I - GWF
* *
(s)GFW
-1
(s) ùû - I + p F GFW
*
}
(s) éë I - GWF
*
(s)G *
) FW (s) ùû .
-1

Equation (1.22) is obtained based on Figure 1.3. The possible transition routes for Ta
are: (1) starts from W, goes through any number (1, 2,¼, ¥) of transitions from W to
F and back to W; (2) starts from F, transfers to W, and then experiences any number
(0,1, , ¥) of transitions from W to F and back to W. So far, all possible transitions
are considered.
Two circumstances should be considered in terms of the comparison between τ
and the length of interval:

i) when b - a < t ,

é a
AY 31 ([a, b]) = I[b- a <t ] ê
ê ò
êë (b-t ) +
fTa (v) exp(QWWt )1W dv

(1.23)
(b-t )+ ù
+
ò0
fTa (v)) exp ( QWW (b - v) ) 1W dv ú .
ú
úû

FIGURE 1.3 Possible sample paths of Ta beginning at W and F, respectively.

Another Random Scribd Document
with Unrelated Content
irregularly disposed and branched, with very numerous clustered blunt
ramules.

Dimensions.—Diameter of the gate 0.07 to 0.09; length of the spines

0.02 to 0.06.

Habitat.—Western Tropical Pacific, Station 225, depth 4475 fathoms.

Subfamily 2. Cortinida, Haeckel.

Definition.—S t e p h a n i d a with typical basal feet (or cortinar feet).

Genus 405. Cortina,[24] n. gen.

Definition.—S t e p h a n i d a with a simple dipleuric or bilateral ring,

bearing at the base three divergent feet (one odd caudal and two paired
lateral feet).

The genus Cortina and the following Stephanium represent together the
small but very important group of Cortinida, differing from the simple
Lithocircida in the possession of three typical basal feet, which are
transmitted to the majority of the Nassellaria by heredity, and produce
their peculiar triradial structure. They appear therefore as a combination
of the simple ring (Zygocircus) with three basal feet (Plagonium). One of
these three divergent feet is the odd caudal foot, opposite to the apical
horn; the two others are the paired lateral or pectoral feet. (On the
probable origin and the typical signification of Cortina compare above,
pp. 891-894.)

1. Cortina tripus, n. sp. (Pl. 83, fig. 9).

Ring ovate, smooth, or with a few short thorns. Apical horn oblique,
curved, simple, smooth, about as long as the sagittal axis of the ring.
Feet divergent, curved, simple, smooth, about as long as the horn; the
caudal foot shorter than the two pectoral feet. Very variable in form and
size.

Dimensions.—Height of the ring 0.06 to 0.09, breadth 0.04 to 0.06;

length of the feet 0.05 to 0.1.

Habitat.—Cosmopolitan; Atlantic, Indian, Pacific, surface and in various

depths.

2. Cortina typus, n. sp. (Pl. 97, fig. 1).

Ring kidney-shaped or nearly semicircular, with revolute vertical dorsal

rod and strongly curved ventral rod; both rods with two pairs of spine-
bunches. Apical horn nearly straight, thorny, longer than the ring,
directed a little obliquely backwards. Feet slightly curved, of equal
length, two to three times as long as the ring; each armed with few
large bunches of curved spines, which are more developed in the caudal
foot than in the two pectoral feet.

Dimensions.—Height of the ring 0.14, breadth 0.08; length of the feet

0.2 to 0.3.

Habitat.—Tropical Atlantic, Station 348, depth 2450 fathoms.

3. Cortina conifera, n. sp.

Ring subcircular, smooth. Apical horn and the three divergent feet of
equal size and similar form, cylindrical, straight, about as long as the
ring, at the distal end thickened, with a dimply cone (similar to
Tripospyris conifera and Tripospyris eucolpa, Pl. 84, figs. 4, 7).
Dimensions.—Height of the ring 0.08, breadth 0.06; length of the feet
0.09.

Habitat.—Central Pacific, Station 268, depth 2900 fathoms.

4. Cortina furcata, n. sp.

Ring elliptical, with three forked horizontal spines (one on the dorsal and
two on the ventral rod). Apical horn and the three divergent feet of
equal size, somewhat longer than the ring, slightly curved, in the distal
half forked.

Dimensions.—Height of the ring 0.07, breadth 0.05; length of the feet

0.09.

Habitat.—South Atlantic, Station 332, depth 2200 fathoms. Fossil in

Barbados.

5. Cortina dendroides, n. sp.

Ring ovate, with six pairs of branched horizontal spines, three on the
dorsal and three on the ventral rod. Apical horn very large, arborescent,
about as long and as broad as the ring. Feet similar to the horn, also
richly branched, of equal size. All the rods and their branches curved,
with prominent distorted edges.

Dimensions.—Height of the ring 0.13 to 0.17, breadth 0.1 to 0.12;

length of the feet 0.15 to 0.2.

Habitat.—Central Pacific, Stations 266 to 274, depth 2350 to 2925

fathoms.
6. Cortina cervina, n. sp. (Pl. 92, fig. 21).

Ring ovate, with four pairs of forked horizontal spines (two dorsal and
two ventral pairs). Apical horn straight and stout, as long as the ring,
with trifid point. Three feet equal, widely divergent, very large, branched
like a deer's antler, with very numerous short and stout, curved and
pointed branches. All rods and branches roundish, without edges.

Dimensions.—Height of the ring 0.1, breadth 0.07; length of the feet

0.12 to 0.16.

Habitat.—Central Pacific, Station 265, depth 2900 fathoms.

Genus 406. Stephanium,[25] n. gen.

Definition.—S t e p h a n i d a with a simple dipleuric or bilateral ring,

bearing at the base four divergent feet (two sagittal and two lateral
feet).

The genus Stephanium differs from the preceding nearly allied Cortina in
the production of four basal feet; the new foot, missing in the latter, is
the anterior or sternal foot. Therefore Stephanium may be regarded as
the archetype of all those Nassellaria in which, on the base of the
sagittal ring, there are developed four typical feet—two sagittal feet (the
posterior caudal and anterior sternal foot) and two lateral feet (right and
left). On the origin of Stephanium compare above, p. 893, &c.

1. Stephanium quadrupes, n. sp. (Pl. 92, fig. 20).

Ring ovate, with three prominent dentate edges and a short pyramidal
apical horn. Four feet all of nearly equal size, about as long as the ring,
also with three thorny edges, in the upper half divergent, in the lower
convergent.

Dimensions.—Height of the ring 0.12, breadth 0.08; length of the feet

0.11 to 0.13.

Habitat.—Central Pacific, Station 266, depth 2750 fathoms.

2. Stephanium tetrapus, n. sp.

Ring elliptical, without edges, thorny, with a stout, thorny apical horn of
the same length. Four feet curved and irregularly branched, divergent,
of different size. The two sagittal feet (the anterior sternal and posterior
caudal) about as long as the ring. The two lateral feet (right and left)
nearly twice as long, more richly branched.

Dimensions.—Height of the ring 0.16, breadth 0.11; length of the feet

0.15 to 0.3.

Habitat.—Central Pacific, Station 272, depth 2600 fathoms. Fossil in

Barbados.

Family XLIX. S e m a n t i d a , n. fam.

Definition.—S t e p h o i d e a with a single vertical ring (the primary

sagittal ring), bearing on its base a horizontal ring (basal or cortinar
ring) with two to four or more basal gates (or cortinar pores).

The family S e m a n t i d a differs from the preceding Stephanida in the

development of a small horizontal ring on the base of the primary
vertical sagittal ring. By the crossing of these two rings a small latticed
basal plate is formed, with one or two pairs of pores; rarely with a
greater number of "basal pores." The production of this characteristic
"basal plate" is of the greatest morphological importance, as the
beginning of the numerous different lattice-formations, which are
differentiated in the great majority of Nassellaria.

In my Prodromus (1881, p. 446) I had enumerated the Semantida with

three genera (Nos. 298 to 300) as a separate subfamily of the
Dyostephida or "Stephoidea biannularia," and characterised these
"Dyostephanida" by the following definition: "Skeleto annulis duobus
composito, qui in duobus planis invicem perpendicularibus jacent; altero
annulo (sagittali) verticali, altero (basali) horizontali." As the names
there given were already employed with another signification, and as
the Zygostephanida (there united with the Dyostephanida) are more
closely related to the Coronida, I now change the names, and propose
to call the family Semantida, expressing by this term the typical
similarity of the skeleton to a signet-ring (Semantis, Semantrum,
Semantidium).

At about the same time, some S t e p h o i d e a of this family were

accurately described by Bütschli (1882, Zeitschr. f. wiss. Zool., vol. xxxvi.
p. 495, Taf. xxxii. figs. 6, 7, 8). He called them Stephanolithis, a name
which Ehrenberg had employed, not for complete shells of Radiolaria,
but for isolated parts of such, and for siliceous fragments of different
skeletons, needles of Sponges, &c. The three species described by
Bütschli represent three different genera of our Semantida, viz.,
Semantis spinescens (with two gates in the basal plate), Semantrum
mülleri (with four gates), and Semantidium haeckelii (with six gates). He
pointed out the great morphological value of the fenestrated basal plate
and its paired gates, as beginnings of numerous other Nassellaria. But
his opinion, that in all S p y r o i d e a and C y r t o i d e a , derived from
these, two pairs of basal gates were constant, was erroneous, nor was
the formation of the first pair naturally explained; he supposed that the
formation of the basal plate begins by development of an odd sagittal
apophysis, arising from the base of the primary sagittal ring. But this
odd sagittal apophysis ("der unpaare mediane Kieselfortsatz c1", loc.
cit., p. 497) is in reality not a primary and essential part of the skeleton,
but secondary and of little morphological value, absent in the majority of
the Semantida and of the other Nassellaria.

We divide our family Semantida into two different subfamilies, which

possibly possess a direct phylogenetic relation to the two subfamilies of
Stephanida:—The Semantiscida have no typical feet, and have arisen
directly from the Lithocircida; the Cortiniscida, however, possess the
three typical basal feet of Cortina, and may therefore be derived directly
from the Cortinida. Since these three cortinar feet are probably identical
with the three primary radial rods of the P l e c t o i d e a , an immediate
affinity also to these Nassellaria is indicated.

The Semantiscida, which do not possess these three basal cortinar feet,
are the simpler forms of the family. The simplest of all, and perhaps the
common ancestral form of the whole family, is Semantis (Pl. 92, figs. 1,
2). It may be derived from Archicircus or Zygocircus by development of
two pairs of horizontal apophyses on its base, around the porochora of
the central capsule. The two rods of each side (right and left), becoming
curved one towards the other, and meeting laterally, form a simple
horizontal gate, and the two paired basal gates together, a horizontal
ring or basal ring, to which the primary sagittal ring is perpendicular. In
the next allied genus, Semantrum (Pl. 92, figs. 3-5), three pairs of
horizontal apophyses are developed, and therefore two pairs of basal
gates produced, an anterior and a posterior. In the third genus,
Semantidium (Pl. 92, figs. 6, 7), three pairs of basal pores or gates are
visible, surrounded and separated by four pairs of horizontal apophyses,
which arise from the base of the sagittal ring. Finally, in Clathrocircus
(Pl. 92, figs. 8-10) the number of apophyses is much increased, and two
parallel rows of pores are developed along the two sides of the sagittal
ring.

The basal plate or the "seal," developed from the base of the primary
sagittal ring or "signet-ring," is therefore a horizontal ring, which
becomes bisected by the latter, and exhibits either one pair of primary
"basal gates" or two or three pairs of these important basal pores, rarely
more. Since these pores possess the greatest morphological value, and
are probably everywhere homologous, we give to them and to the
separating apophyses certain names, and call the anterior pair of gates,
"jugular pores" (i in our figures, the pair I of Bütschli); the middle
(usually the largest) pair, "cardinal pores" (k in our figures, the pair II of
Bütschli), and the posterior, smaller pair, "cervical pores," l. The typical
pairs of rods, by the union of which these basal pores arise, are the
following:—(1) the clavicular or furcular rods, f, the first pair (rods e1 of
Bütschli), (2) the coracal rods, e, between the jugular and cardinal pores
(rods e of Bütschli), (3) the scapular rods, g, between the cardinal and
cervical pores (rods e2 of Bütschli), (4) the cervical rods, the fourth pair
of apophyses, the most posterior, h. Bütschli supposes that the
topographical succession of the three typical pairs of basal pores is also
the chronological succession, the jugular being formed first, the cardinal
second and the cervical pores third but it seems that this succession is
often altered and that the cardinal pores (the largest), appear first, the
jugular pores (in front of them) second and the cervical pores third (or
perhaps sometimes in the inverse succession).

The Cortiniscida, the second subfamily of Semantida, differ from the

preceding Semantiscida in the possession of three typical basal feet,
which are probably inherited from the Cortinida, and perhaps indirectly
from the Plagonida (Plagoniscus). The simplest and the most important
form of this second family is Cortiniscus (Pl. 92, figs. 11-13), differing
from its probable ancestral form, Cortina, in the development of a basal
ring, produced by horizontal union of the basal apophyses. The three
typical feet are the same as in all triradiate Nassellaria, two paired
pectoral feet (p1, p11) diverging on the anterior base of the sagittal ring,
whilst the odd caudal foot (c) lies on its posterior base in the sagittal
plane and appears as a basal prolongation of the dorsal rod of the
sagittal ring; the upper part of the dorsal rod is usually prolonged into
an ascending apical horn. In Stephaniscus (Pl. 92, figs. 14, 15) four
basal feet are visible, an anterior or sternal foot (z) being added as a
prolongation of the basal rod of the ring, opposite to the caudal foot.
Finally, Semantiscus (Pl. 92, figs. 16-18) is distinguished by the
possession of six divergent basal feet, probably identical with those of all
six-radiate Nassellaria; three of these may be regarded as primary and
perradial, the odd caudal and the paired pectoral feet; the three other
intercalated as secondary or interradial feet, the odd sternal (z) and the
paired tergal feet (t1, t11).

The basal plate of these Cortiniscida exhibits the same important

differences as in the preceding Semantiscida, either one, or two, or
three pairs of basal gates being developed. But there occur also in some
species (mainly in Cortiniscus) only three basal gates, an odd anterior
(between the two pectoral feet and a connecting horizontal bar), and
two paired posterior (between the two pectoral and the odd caudal
foot). It requires further accurate researches to solve the important
problem, what the true homologies of these typical basal pores and the
separating bars are in the different genera of Semantida. In Semantiscus
there are three pairs of basal pores in the horizontal seal, corresponding
to those of Semantidium and the radial rods or bars between, these are
the basal parts of the six radial feet; therefore the odd caudal foot (t)
seems to be the posterior, and the odd sternal foot (z) the anterior
prolongation of the basal part of the primary sagittal ring; the two
paired anterior or pectoral feet (p1, p11) the prolongations of the coracal
rods (e) and the two paired posterior or tergal feet (t1, t11), the
prolongations of the scapular rods (g). But it is not yet certain whether
these six radial feet and the separating gates of the basal plate are all
the same and truly homologous in all six-radiate Nassellaria. In every
case the comparative study of the Semantida is of the highest value for
the accurate knowledge of the Monopylea.

Synopsis of the Genera of Semantida.

 Only
two, Two basal
four, pores,
407. Semantis.
or six
I. Subfamily basal
Semantiscida. Four basal
pores
Basal ring pores,
(no 408. Semantrum.
without dorsal
typical, and Six basal
regularly ventral pores,
disposed pores). 409. Semantidium.
basal feet (no
cortinar feet). Besides four basal
pores, also apical
pores or dorsal and
ventral pores along
the whole ring, 410. Clathrocircus.
One odd
caudal
Three foot and
basal two
feet. paired
lateral
feet, 411. Cortiniscus.
II. Subfamily
Cortiniscida. Two
Basal ring with sagittal
typical, Four and two
regularly basal lateral
disposed feet. or
basal feet pectoral
(cortinar feet, 412. Stephaniscus.
feet).
Two
sagittal,
two
Six basal
pectoral,
feet.
and two
tergal
feet, 413. Semantiscus.

Subfamily 1. Semantiscida, Haeckel.

Definition.—S e m a n t i d a without typical basal feet or cortinar feet.

Genus 407. Semantis,[26] n. gen.

Definition.—S e m a n t i d a with two basal pores (or jugular pores),

without typical basal feet.

The genus Semantis, the most primitive and the oldest of the
Semantida, is of the greatest morphological interest, as the first form of
S t e p h o i d e a which produces gates or pores by communicating
branches, and therefore the probable ancestral form not only of this
family, but of the greater number of all S t e p h o i d e a , and perhaps
even of all S p y r o i d e a and C y r t o i d e a . Semantis arises from
Archicircus by the production of two pairs of lateral branches from the
basilar rod of the sagittal ring, one anterior pair of clavicular rods, and
one posterior pair of coracal rods. By junction of the clavicular and
coracal rod on each side arises a left and a right pore, the "jugular pore
or jugular gate."

1. Semantis biforis, n. sp. (Pl. 92, fig. 2).

Sagittal ring obliquely ovate, nearly trapezoidal, with six pairs of short
mammillated knobs or branches; dorsal rod straight, vertical, with two
pairs of knobs, ventral rod strongly convex, also with two pairs of knobs;
two other pairs in the apical rod. Basilar rod horizontal, straight. Basal
ring with six pairs of similar knobs, three on each side. Basal gates ovate
or nearly triangular.

Dimensions.—Height of the sagittal ring 0.09, breadth 0.07.

Habitat.—Central Pacific, Stations 265 to 268, depth 2700 to 2900

fathoms.
2. Semantis distoma, n. sp.

Sagittal and basal ring nearly of the same form as in the preceding
species, but half as thick and armed with numerous branched spines
which are about half as long as the diameter of the main gate, and with
thin and curved pointed branches.

Dimensions.—Height of the sagittal ring 0.11, breadth 0.08.

Habitat.—North Pacific, Station 244, depth 2900 fathoms.

3. Semantis dipyla, n. sp.

Sagittal ring kidney-shaped, very thick, with six pairs of short, thorny, or
irregularly tuberculated knobs (two dorsal, two apical, and two ventral
pairs). Basal ring smooth, scarcely half as thick, horizontal, with two
semicircular basal gates, about half as broad as the main gate. From the
two opposite lateral corners of the basal ring two slender upwardly
curved spines arise, resembling the basal part of a commencing frontal
ring.

Dimensions.—Height of the sagittal ring 0.08, breadth 0.05.

Habitat.—Western Tropical Pacific, Station 225, depth 4475 fathoms.

4. Semantis sigillum, n. sp. (Pl. 92, fig. 1).

Sagittal ring elliptical, with four pairs of stout arborescent spines (two
apical and two equatorial pairs), which are irregularly branched and
forked, with numerous thin lateral branches. Apex with a short conical
trifid vertical horn. Basal ring thorny, with obliquely descending slightly
curved bars.
Dimensions.—Height of the sagittal ring 0.12, breadth 0.09.

Habitat.—South Pacific, Station 300, depth 1375 fathoms.

5. Semantis distephanus, n. sp. (Pl. 83, fig. 3).

Sagittal ring thin, semicircular, smaller than the thin basal ring, which
exhibits two semicircular gates. Both rings are armed with numerous
small spines of equal size, which on the former are arranged in two, on
the latter in three regular rows. (The basal rod of the sagittal ring,
separating the two basal gates, in fig. 3 is, by mistake, not distinctly
enough drawn.)

Dimensions.—Height of the sagittal ring 0.06, breadth 0.08.

Habitat.—Tropical Atlantic, Station 348, surface.

6. Semantis spinescens, Haeckel.

Stephanolithis spinescens, Ehrenberg, 1875, Abhandl. d. k. Akad. d. Wiss. Berlin,

p.160, Taf. i. fig. 29.
Stephanolithis spinescens, Bütschli, 1882, Zeitschr. f. wiss. Zool., vol. xxxvi. p.
497, Taf. xxxii. figs. 7a, 7b.

Sagittal ring ovate with four pairs of thin, irregularly branched spines,
two apical pairs, one on the straight dorsal rod and one on the curved
ventral rod. Basilar rod with a posterior and an anterior forked rod
(commencing caudal and sternal foot). Basal ring square, with two
lateral spines, and two triangular gates scarcely one-third as broad as
the ring-gate. An internal ascending procolumna (rod c1 in the figure of
Bütschli) connects the basal and ventral rods of the sagittal ring.

Dimensions.—Height of the sagittal ring 0.1, breadth 0.07.

Habitat.—Fossil in Barbados.

Genus 408. Semantrum,[27] n. gen.

Definition.—S e m a n t i d a with four basal pores (two anterior jugular

and two posterior cardinal pores), without typical basal feet.

The genus Semantrum, one of the most important of the Nassellaria,

arises from the preceding Semantis by duplication of the two basal
gates. Behind the pair of coracal rods there arises from the basilar rod of
the sagittal ring a third pair of lateral horizontal branches, the scapular
rods. These become connected with the coracal rods on each side, and
so produce a second posterior pair of basal pores, the "cardinal gates."
These are constantly larger than the anterior "jugular gates." Therefore
the vertical ring of Semantrum possesses a horizontal basal ring with
four very characteristic gates, enclosed by three pairs of lateral curved
and connected branches, and these become transmitted by heredity to
the majority of the Nassellaria.

1. Semantrum quadrifore, n. sp. (Pl. 92, fig. 5).

Sagittal ring subcircular or ovate, with three edges and four sagittal
forked spines on the odd edge (two dorsal and two ventral spines).
Basal ring decagonal, with ten simple or forked spines on the ten
corners. Jugular gates tetragonal. Cardinal gates pentagonal.

Dimensions.—Height of the sagittal ring 0.09, breadth 0.07.

Habitat.—Tropical Atlantic, Station 351, surface.

2. Semantrum tetrastoma, n. sp. (Pl. 92, fig. 3).

Sagittal ring semicircular, thorny, without edges. Basal ring tetragonal or
nearly cordate, with a small anterior and a large posterior bow-shaped
incision; on the lateral edges with numerous irregular thorns. Jugular
gates pear-shaped. Cardinal gates nearly kidney-shaped.

Dimensions.—Height of the sagittal ring 0.1 to 0.14, breadth 0.07 to

0.09.

Habitat.—Central Pacific, Stations 265 to 268, depth 2700 to 2900

fathoms.

3. Semantrum tetrapylum, n. sp.

Sagittal ring elliptical, tuberculate. Basal ring trapezoidal, also with

roundish thick rods and small irregular tubercles on the margin. Jugular
gates ovate, two-thirds as broad as the triangular cardinal gates.

Dimensions.—Height of the sagittal ring 0.08, breadth 0.05.

Habitat.—North Pacific, Station 253, depth 3125 fathoms.

4. Semantrum mülleri, Haeckel.

Stephanolithis mülleri, Bütschli, 1882, Zeitschr. f. wiss. Zool., vol. xxxvi pp. 499,
539, Taf. xxxii. figs. 8a, 8b, 8c.

Sagittal ring nearly semicircular or obliquely ovate, thorny. Basal ring

also nearly semicircular, with two slight sagittal incisions (one anterior
and one posterior), on the lateral edges with numerous small thorns.
Jugular pores (pair I of Bütschli) pear-shaped. Cardinal pores (pair II of
Bütschli) triangular. Sometimes (but not constantly) a pair of cervical
bows connects the scapular bars with the subvertical dorsal rod of the
sagittal ring.
Dimensions.—Height of the sagittal ring 0.09, breadth 0.06.

Habitat.—Fossil in Barbados.

5. Semantrum sphragisma, n. sp.

Sagittal ring circular, smooth. Basal ring also nearly circular and smooth;
its four basal gates of nearly equal size, elliptical or subcircular; the
jugular pores scarcely smaller than the cardinal pores. All rods smooth,
cylindrical, without edges and thorns.

Dimensions.—Height of the sagittal ring 0.08, breadth 0.07.

Habitat.—Tropical Atlantic, Station 347, depth 2250 fathoms; also fossil

in Barbados.

6. Semantrum bütschlii, n. sp.

Sagittal ring triangular, with vertical straight ascending dorsal rod,

horizontal straight basilar rod and obliquely ascending, slightly curved
ventral rod; all rods nearly cylindrical, with irregular branched thorns
and distorted edges. Jugular pores triangular, scarcely half as broad as
the large semicircular cardinal pores.

Dimensions.—Height of the sagittal ring 0.11, breadth 0.08.

Habitat.—Fossil in Barbados.

7. Semantrum signarium, n. sp. (Pl. 92, fig. 4).

Sagittal ring obliquely ovate, thorny. Basal ring kidney-shaped or nearly
pentagonal, with broad concave dorsal incision and five larger branched
spines on the five corners; between them smaller thorns. Jugular gates
pear-shaped, scarcely half as broad as the triangular cardinal gates.

Dimensions.—Height of the sagittal ring 0.12, breadth 0.09.

Habitat.—Central Pacific, Station 271, depth 2425 fathoms.

Genus 409. Semantidium,[28] n. gen.

Definition.—S e m a n t i d a with six basal pores (two anterior jugular,

two middle cardinal, and two posterior cervical pores), without typical
basal feet.

The genus Semantidium exhibits a further developmental stage of the

basal plate of the Semantida. Whilst this plate in Semantis possesses
one pair of basal pores, and in Semantrum two pairs, here in
Semantidium it has three pairs. The new third pair is produced on the
dorsal edge of the basal plate, which becomes connected with the basal
part of the dorsal rod of the sagittal ring by a pair of cervical rods. The
middle pair of pores (the cardinal) are always larger than the anterior
(jugular) and the posterior (cervical pores). The same form of basal
plate is preserved in numerous S p y r o i d e a and C y r t o i d e a , as a
"cortinar septum with six collar pores," (e.g., Pl. 53, fig. 18).

1. Semantidium hexastoma, n. sp. (Pl. 92, fig. 6).

Sagittal ring ovate, thorny. Basal ring rhombic or nearly square, with
four short conical descending spines on the four prominent edges (two
sagittal and two lateral); between them numerous smaller irregular
thorns. Jugular and cervical gates nearly equal, ovate, half as broad as
the triangular cardinal gates between them.
Dimensions.—Height of the sagittal ring 0.14, breadth 0.1.

Habitat.—Central Pacific, Station 265, depth 2900 fathoms.

2. Semantidium sexangulum, n. sp.

Sagittal ring ovate, with six pairs of short branched thorns (two apical,
two dorsal, and two ventral). Basal ring hexagonal, with three pairs of
short branched thorns on the six corners (one sagittal pair, one posterior
and one anterior). All six gates of the basal plate triangular, the jugular
and cervical a little smaller than the cardinal gates.

Dimensions.—Height of the sagittal ring 0.1, breadth 0.07.

Habitat.—Tropical Atlantic, Station 348, depth 2450 fathoms.

3. Semantidium haeckelii, Bütschli.

 Stephanolithis Haeckelii, Bütschli, 1882, Zeitschr. f. wiss. Zool., vol. xxxvi. pp. 499, 538, Taf. xxxii. figs. 6a,
6 b.

Sagittal ring elliptical, with three pairs of short horizontal branched spines, one apical and two
equatorial pairs (one dorsal and one ventral). Basal ring roundish hexagonal, with numerous
short thorns on the margin. Jugular pores ovate, about half as broad as the ovate cardinal
pores and twice as broad as the small cervical pores.

Dimensions.—Height of the sagittal ring 0.08, breadth 0.06.

Habitat.—Fossil in Barbados.

4. Semantidium signatorium, n. sp. (Pl. 92, fig. 7).

Sagittal ring semicircular, thorny; basal ring pentagonal, with short spines on the margin and
five stronger thorny spines on the five corners. Jugular pores ovate, smaller than the triangular
cervical pores. Cardinal pores two to three times as large as each of the former, pentagonal.

Dimensions.—Height of the sagittal ring 0.08, breadth 0.12.

Habitat.—North Pacific, Station 341, depth 2300 fathoms.

Genus 410. Clathrocircus,[29] Haeckel, 1881, Prodromus, p. 447.

Definition.—S e m a n t i d a with a variable number of pores on the apical and the basal part of
the ring, symmetrically arranged, without typical basal feet.

The genus Clathrocircus comprises those Semantida in which the sagittal ring bears not only
basal pores (as in the three preceding genera) but also apical pores (on the opposite pole of
the main axis), or a variable number of pores along the whole ring. All these pores are
symmetrically arranged in pairs. In the simplest form there are only two apical pores opposite
to four basal pores, whilst in the highest state of development the whole ring bears two
complete circles of pores. At both poles of the transverse axis two large lateral gates remain
open. If these become closed by lattice-work, Clathrocircus passes over into Dictyospyris.

1. Clathrocircus hexaporus, n. sp.

Sagittal ring circular, smooth, with three pairs of pores (one apical and two basal). The two
apical pores are triangular with two lateral spines, and correspond to the two mitral gates of the
Tympanida. The four basal pores form a hexagonal basal plate, with six lateral spines, and
correspond to the four basal pores of Semantrum; the two jugular are pear-shaped, and half as
broad as the two triangular cardinal pores.
Dimensions.—Height of the sagittal ring 0.08, breadth 0.06.

Habitat.—Cosmopolitan; Atlantic, Pacific, many Stations, surface.

2. Clathrocircus octoporus, n. sp.

Sagittal ring elliptical, smooth, with four pairs of pores (two apical and two basal). The four
apical pores are of nearly equal size, subcircular, and form a regular cross around the apical
pole. The four basal pores are ovate, and form a quadrangular plate, armed with marginal
thorns; the two jugular pores are somewhat smaller than the two cardinal pores.

Dimensions.—Height of the sagittal ring 0.11, breadth 0.08.

Habitat.—Tropical Atlantic, Station 348, depth 2450 fathoms.

3. Clathrocircus decaporus, n. sp.

Sagittal ring ovate, thorny, with five pairs of pores (two apical and three basal). The four apical
pores are roundish and form a cross. The six basal pores are triangular and form a hexagonal
plate with thorny margin; the two jugular pores are somewhat larger than the two cervical, and
half as broad as the two cardinal pores.

Dimensions.—Height of the sagittal ring 0.09, breadth 0.07.

Habitat.—North Pacific, Station 256, depth 2950 fathoms.

4. Clathrocircus stapedius, n. sp. (Pl. 92, fig. 8).

Sagittal ring ovate, smooth, with six pairs of pores (two apical and four basal). The four apical
pores are of nearly equal size and form a quadrangular mitral plate with two large lateral horns.
The eight basal pores are of very different size (the four central far larger than the two anterior
and the two posterior), and form a hexagonal basal plate, also with two large lateral horns. The
four branched lateral horns (two upper and two lower) form together an incomplete frontal
ring.

Dimensions.—Height of the sagittal ring 0.1, breadth 0.08.

Habitat.—Central Pacific, Station 271, depth 2425 fathoms.

5. Clathrocircus dictyospyris, n. sp. (Pl. 92, fig. 9).

Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com